PS: I am not sure about where your reasoning is coming from. What 26.2% at 15 fps means is that 26.2% of the occurrences of the target behaviour (= not necessarily matching a given frame) succeeded. You cannot extrapolate this information to different conditions, like higher/lower number of frames per second.
I took the information from the paper itself, more specifically from the following part: "For example, when
training on the KITTI pedestrian dataset [18], the best known average miss rate is 61.2%,
whereas when training on INRIA [10], the average miss rate is 50.2% [6]. Both miss rates are
much higher than 31.1% of our method"
As I understand it, a miss rate of 31.1% on a given dataset means that 31.1% of the tested attempts failed.
Idea 1: the fact that accounting for the eventuality of not hitting pedestrians (or any other being/thing) is "one of the unexciting but vital bits of technology" when talking about an autonomous car provides a quite accurate summary about what a big proportion of AI-focused approaches are about. Lots buzz (= exciting technological break-troughs) and not actually-working results (= unexciting technical bits avoiding the big idea to work at all). And this is not just what the OP thinks; Google has been testing autonomous cars for some years already without having still tackled such a secondary(?!) issue.
Idea 2: after quickly skimming through this paper, it seems that the new much-more-accurate algorithm still misses 30% of cases. For me, hurting (even killing) 3 out of 10 pedestrians still sounds quite bad. Additionally, we are talking about their training dataset whose exact complexity is not too clear. For example: what about a kid suddenly crossing the street?, how good is this algorithm at differentiating between persons and similar shapes (human-like advertisement)?, how does it behave in poor-visibility conditions?, etc.
I am completely aware about the tremendous difficulty associated with accomplishing the expected goal and the outputs so far seem promising. But why are they implying that something is almost done, when quite a few basic problems haven't still been tackled?
Sorry to read about your problems, but all this has nothing to do with me: I rarely get angry; and when I do, I would rarely make decisions; and even in case of making decisions, I would always take full responsibility for my actions (I take responsibility for what I did even when being a kid) and eventually re-decide anything.
I am a very sensible guy who has traditionally been surrounded by very sensible people. During the last years, I have been forced to deal with not-too-sensible individuals, the kind of people with whom your statement "when everybody is yelling then nobody is listening" makes sense at all.
As explained in my reply to the comment above, the referred aggressive emails were a gesture of generosity (not of angriness); equivalently to what I have done in quite a few other contexts. But also as explained, I am quite tired now but also sure that, in some situations, I should be more egoist.
This is how I (used to) see things: 1. You are doing something which can be criticised by others without being truly able to defend your position. Example: keep using someone else's email address after a nice request from its rightful owner. 2. You are not fully aware about the aforementioned situation for whatever reason. 3. Someone replies you quite aggressively but by providing lots of useful information. 4 (what I was expecting). Think carefully about what you are exactly doing, understand that you cannot defend your position, learn the lesson and move on. 4 (what I have seen lots of times). People obsessively repeating the same actions by completely altering the original intention. For example: keep sending the same offers to a person who has let very clear his position regarding the product/company.
Among grow-ups I don't have to take care of the education (and/or account for the complexes) of each single person I met. My aforementioned previous (aggressive) attempts were quite generous gestures, which some people might have taken as a help to understand that they were acting wrong.
Your whole speech, although undoubtedly nice and coherent in appearance, actually hides ideas on the lines of "never say you did it wrong", "everyone is right" or similar; the kind of trendy educational attitudes outputting spoiled children with neither boundaries nor respect for others. If you did something wrong, accept the consequences; and stop looking for excuses to somehow compensate your actions.
I have always been an extremely adaptable and understanding person, always thinking that we all are identical and trying to help my fellow men to see their errors. I am very tired now and I have confirmed that generosity can only be delivered within properly-understanding contexts; that’s why I only want to deal with properly understanding individuals. With all the other people, I will rely on the referred “better ignoring them" attitude, much more egoist than my previous behaviour but proven required.
Thanks for having shared so profound thoughts. I do understand the intrinsic limitations of your actions and will overlook the tremendous irony underlying your whole post. The fact that you wanted to ignore what I wrote has indeed hurt my feelings. I hope that other person will inflict you approximately the same pain which I am feeling right now.
PS: I will not say anything about your meta references.
I used to think that ignoring someone's email was the worst you could do to that person. In fact, I used to write lengthy replies with a quite aggressive tone to people doing really stupid things; having ignored them would have been 1000 times easier for me. I wasn't angry at all (I have rarely done anything on the lines of e-venting), but didn't find any other way to communicate with people not able to understand the simplest idea (e.g., spammer continue spamming after you say that are not interested in these products).
The results were really bad. I came to the sad conclusion that there are quite a few people with not much self-respect, who just look for any excuse to blow everything out of proportion and evade all responsibility for their actions (e.g., "why are you screaming to me?", "because you screamed to me before", "I don't care! You cannot scream to me!").
Now I am always using the easiest alternative (although I still think that it is the most aggressive and disrespectful one): I plainly ignore people showing nonsensical behaviours.
Out of two jokes, rightly choosing the funnier one 64% of the times can be considered as almost random; mainly when analysing a pretty small dataset (what is presumably the case here). Bear in mind than reaching the ideal 50% value might be difficult even for a random number generator and lots of attempts might be required.
The most logical way to show results on these lines is by including a better description of the testing conditions (e.g., X samples, by considering Y of them to be negligible because this and that factors and Z to have these peculiarities).
Just as a blind number, 55.8% sounds horrible and 64% sounds still pretty bad. If I was a New Yorker editor, I wouldn't ever trust in such a tool to perform my work. And this is precisely the whole point here: not considering that 64% is good enough because of not having anything better (or because of being a good result for a computer), but because of being actually trustworthy. If you cannot reach such a stage, better keep working and don’t show as a victory what has no real value.
The situation you describe is different to what this algorithm has to deal with (just funny vs. not funny). In any case a 55% success rate would be equally bad in both scenarios.
For example: if out of 10 funny cases, the algorithm concludes that 5 are funny and the other 5 are not; it is irrelevant the fact of having just one not-funny option or as many additional ones as required (interesting, insightful...); the algorithm would fail in a very relevant proportion of the cases. And its failure would be very similar to the one output by a random decision maker (whose theoretical error is precisely 50% when deciding whether something is in some way or not).
Anything below 75% converts almost-understanding into random behaviour in most of the situations. What would you think about a person getting only half of the jokes you tell? Better: about a person just understanding that your sentences are expected to be forming a joke (properly getting the joke is a completely different story).
Using the New Yorker jokes to train the algorithm?! About 55.8% of the time the humans agree with the captions the AI selects (55.8% of people finding funny the New Yorker jokes, who are less than 10% of all the humans in the world)?!
It seems to me that this is a new generator of almost-random answers with a surprisingly high cost, because their not-particularly-knowledgeable new owners are glad to pay whatever is required to feel that they are part of the AI revolution.
CLARIFICATION: Review (any of the aforementioned versions) is a comedy TV show, a mockumentary, where a guy rates the craziest situations (e.g., starting a cult). It is one of the funniest shows I have ever seen (both versions, although I prefer the American one). Some people might think that this clarification is not required, mainly after having added a link to a video clearly showing that it is a comedy show. But I have met lots of people lately with serious proper-understanding problems and am quite tired or their noisy nonsense. This is the reason why now I always make completely sure that every single bit is perfectly clear to anyone.
I will think twice (and thrice) before posting a new not-too-direct sarcastic remark here just to avoid this kind of not-funny-at-all clarifications.
Walking on water!? I don't need anything else, It is undoubtedly the new Messiah. This is excellent news for me, as far as I was planning to start a cult anyway.
To confirm how serious I am about all this and to celebrate that the second season of Review (with Forrest MacNeil) started yesterday, here you have a worth-watching video: https://www.youtube.com/watch?v=jRZIIms7MCU (Starting a cult. Review with Myles Barlow, the Australian prequel of Review).
It was certainly an exclusive. In fact, Mega has defended itself just some minutes ago, by tweeting a reference to this article:
https://torrentfreak.com/kim-d...
Being a billionaire nowadays is certainly hard; in fact, this is the main reason why I am poor. There is a tremendous peer pressure because all the other billionaires are donating lots of money to good causes. I imagine what a billionaire might be thinking while choosing his defining good cause: “This one is not cool enough”; “Bill is already taking care of this other one”; “Is that one a scam? Will keep it on hold for the time being”; “This one is too expensive and spoiling my kids is not precisely cheap”. Billionaires certainly miss simpler times, when they were keeping all their money to themselves. Poor men/women!
Sorry, I answered this one also without thinking to much and right after completing my long answer to your comment above. Better read firstly my answer to your last message below (my tone is more relaxed).
Fusion power is defined as the amount of energy per time released by fusion reactions in the experiment,
Nice thing to know. So nothing to do with a standard nuclear (fission) plant, where the power is measured by the electricity being generated. That is: your fusion "reactors" (or test reactors or expensive toys) are measuring the generated power in a completely different way than what is standard (and actually relevant; as far as the whole point of fusion is generating electricity). Is not this a somehow-misleading way to refer to very important concepts? on one hand, you are using the same word (reactor) to refer to two different realities; on another hand, you are measuring different variables but using the same name (generated power).
Besides, your original post said, " macroscopic-relevant fusion reaction," when I think most would consider MW of reactions to be quite macroscopic
Lots of misunderstandings over here, that's why I thought that writing the aforementioned clear-enough summary was required. With macroscopic relevant I mean something which has an actual effect at the macroscopic level, like heating a cup of coffee during 5 minutes. Playing around with atoms is required, but it is just a pre-step. For example, in fission: you have to firstly have control over the fission reactions (start and terminate them at will), then convert the outputted energy in what you want (i.e., source of heat able to keep the temperature in a water tank with certain dimensions at around 100C) and finally make sure that the whole process is stable and regular enough (like a switch you can set on and off to get what you want). With macroscopic effects I meant having a preliminary version (needing whatever crazily big resources are required) of a system capable to perform all the steps until the end, that is: actually generating a usable source of heat. Making sure that the atoms behave as you want is one of the intermediate steps and what I call "microscopic level" as far as my cup of coffee can still not be heated.
A practical fusion reactor does not require a completely self-sustaining chain reaction to run,
Such a statement sounds very compatible with the fusion theory and reality (= labs completely focused on making sure that the first steps work well enough, still very far away from caring about the actual applicability of these first steps), but unfortunately extremely incompatible with the actual reality of electricity generation. As explained in my previous point, what a fusion reactor is expected to do (at least, ITER the first real fusion reactor) is applying the Rankine cycle (the one used in all the fission plants) which expects a regular source of heat at certain temperature. Regular means over months (I think that the standard fission plants are stopped just once or perhaps twice a year just for security reasons; what means that the chain reactions are automatically happening during all that time); you cannot have a power plant generating electricity at irregular intervals. I think that ITER expects to reach more or less self-sustained chain reactions, equivalently to what happens in fission reactors, basically because this is what is required: a regular (over months) source of heat keeping the temperature of the water in a tank within a very narrow range. I am afraid that you cannot get that with "shots" of extremely hot reactions, which can be repeated once every x minutes. Or even in case of getting such a situation (still quite far from that anyway), the resulting system would be extremely inefficient in comparison with any other fuel. Note that additionally to requiring sun-like temperatures to start the process, you have to spend resources on containing the hot plasma (to eventually refrigerate it) and to convert the high temperatures into the target one of around 100C. The only way to make such a crazily inefficient system to be more or less acceptable would be by relying on self-sustained reac
You know about this thing of time zones and time difference, don't you? Is this the kind of thing that a person (well... presumably; an anonymous coward can be many things:) I am joking, don't get angry just for this) with background in plasma physics can know? I am in Spain; the last time I checked this comment was 10pm my time; in any case, why should I come here to answer you immediately?! Why you think that anything you say is relevant? I am systematically defining all what you are saying as nonsense and telling you that "will stop answering you because doesn't make any sense to me"; isn't this enough to understand that your opinion is not important to me? This is kind of funny, because now I am living in a small town in the middle of nowhere, where people show an equivalent behaviour: they seem to think that their opinions have an intrinsic value no matter how many times you say them "what you are saying do not make any sense to me". They are also very nervous and impatient. Anyway...
(Seriously, last set of answers to you because you are not the kind of person I want to discuss with).
Funny how you demand references from other people, but act like we should depend on your above average background as you've never provided any references to your claim
What is the exact problem between the reality (= the current conversation which you can read as many times as you wish) and the weird conclusions you draw and the pointless replies you write? At what point have I demand anything?! You were the one doubting about my knowledge and I have told you about my knowledge. I have merely forwarded such a request back. I have ever doubted about anything you said (despite the subtle difference that I am a logged-clearly-identified user and you are discussing as an anonymous coward): you didn’t say plainly anything. You are now doubting about what I am saying?! Seriously, what is wrong with you?
you were given the exact name of a major 15+ year experiment
Have you even gone to the University? For what would an engineer want to know the names of experiments about something of no practical use (not for engineering, not for anyone)? In Industrial Engineering (in Spain, 10 years ago), I got my BS in Mechanical Engineering without hearing a word about nuclear fusion (not even fission). When studying for my MS (4th & 5th), precisely specialising in energy (i.e., power plants and engines) and despite having quite few fission-, power-plant- and electricity-generation-related subjects I didn’t hear a word about fusion, other than side comments on the lines of “never has been done”, “impossible dream”, “during over the last 40 years, many people have promised fusion in 10 years times”, etc. The only time when I studied something related to abstract physics theories was in an introductory subject (which was precisely criticised because of being unnecessarily abstract and useless; BTW I got 9.5 over 10) mentioning Relativity and Quantum Mechanics (BTW perhaps you want to know my opinion about the first one, I call it: “Critical Analysis of the Main Premises of Special Relativity: Lorentz & Minkowski”). Why? Because engineers do not care about these things, because the actual applicability of these theories is none (please, refer to the aforementioned critical analysis)! I wasn't here talking as an expert in fusion, because I am not (I don't even believe that it is possible). I was talking on account of my knowledge (the one I got in the University and in some later researches because I like the field; but not fusion; nuclear engineering and fission and actually-applicable physics); was sharing what I think that has been done and can be done. I said it very clearly: correct me if I am wrong, but support your claims with reliable references. Why? Because I have a solid enough knowledge to know that certain things haven't happen yet (and what is required to make them happen is so complex and so expensive that are extremely unlikely to happen within the quite a few next years). Or by using other words: I don’t need to be an expert magician to discuss about magic existence. In fact, most of people defending that magic does not exist are not magicians; why would be otherwise? And you are blaming me for not knowing magic? No, I don’t know magic, why should I?
(e.g. JET, EAST, DIII-D, KSTAR, Tore Supra, ASDEX, among many more), plus a couple dozen medium scale projects and other large scale projects of competing design
I talked about this in my comment below (which I think you have also answered; will look at it right after finishing this comment). These are experiments at universities. You can call them reactors because this is the usual name for the container where (nuclear) reactions happen. But you shouldn’t use this term directly (your beloved Wikipedia uses the expression “test reactor”; everything is fine as far as t
I am including below these lines a comment I wrote above as a reply to another person (providing links to various fusion reactors). I think that these ideas are also helpful here to adequately explain my exact position on the fusion-reactors front:
-----------------
To avoid future misunderstandings, here you have a set of more detailed (perhaps too evident at some point; but will surely avoid misterpretations) ideas:
- What we want fusion for? For somehow emulating what the current fission reactors (logically by bearing in mind the multiple differences between both phenomena) are doing; more specifically, to emulate the stable (and not-too-high-temperature) heat generation outputted by the fission reactions. Why the fission reactors have to generate stable (and not-too-high-temperature) heat? Because all the fission-based power plants (equivalently to what is expected to happen with the future(?) fusion ones) are based on a very simple but perfectly working idea: the electricity is generated from the rotation of a turbine; this turbine has to be regularly rotating for as long periods as possible (because we need all the possible electricity and more); this rotation is achieved by boiling water, phenomenon which happens only its temperature is within certain pretty small range (a bit below/above 100C). The current nuclear (fission) plants get this stable source of heat (with not-too-high-temperature) by taking advantage from what the fission reactions generate. This generated temperature is quite low and that’s why it represents a good source of energy on this context (i.e., a good waiter-boiler). You also don’t need too much energy to start the fission reactions; in fact: you don’t need to start them too often (almost never), this is the beauty of the chain fission reactions and that’s why this form of energy (independently upon the associated dangers) is very efficient and that’s why the nuclear fission plants are the most powerful ones.
- The fusion reactors are expected to apply exactly the same ideas than the fission ones; because the fusion reactions are also expected to generate a regular source of heat (which is basically what we want). The problem is that this time, starting the process (and maintaining it) is not so easy. Additionally the temperature of the generated heat is so high that provokes lots of problems. Creating a chain reaction in fission is very easy; you can see it as a pool (billiard) game: you through a ball which hits another one, which hits the wall, which hits another one After the process is started, it will continue for really long. In fact, without chain reactions nuclear energy would make any sense (= couldn’t considered a form of energy) as far as the heat generated in just one event (fission/fusion) is not too relevant. When you say that fusion hasn’t reached the break-even or that you have put in x and it can only generate x-1, you are not defining a form of energy, but a form of consumption (equivalently to the water which is expected to be boiled).
- Thus, when I talk about fusion I understand a form of energy capable to deliver what is expected, that is: generating heat (at an adequate temperature) for a long enough period of time (to boil water which will then generate electricity). If you are able to fuse two nuclei, excellent; if these first two nuclei generate enough heat to automatically provoke the fusion of the neighbour ones (and so on and so forth), excellent too; if you need lots of energy to perform these fusions but you create a properly working framework, excellent. But don’t call it form of energy if it does not behave as what is traditionally called form of energy (in this context: stable water boiler). If after years (and years and years) the maximum you get (after having spent imaginably high amounts of money) is some nuclei to fuse, a system able to generate and contain the disproportionately high required temperatures, and even get some chain reactions for some seconds Be very proud of
First of all: thanks for your patience and for making the (small) effort of putting together these ideas. I do recognise that have learned some things which I didn’t know (e.g., the magnetic container proposed by the ITER has already been tested in many different situations, although just during some seconds).
Other than that, I am afraid that my ideas haven't changed appreciably (at least, not in the I-was-wrong direction). I think that there are quite a few not-completely-clear-to-everyone points which provoke systematic misunderstandings. Thus, to reward your effort (, to prove that your interpretation of my behaviour was completely wrong) and to avoid future misunderstandings, here you have a set of more detailed (perhaps too evident at some point; but will surely avoid misterpretations) ideas:
- What we want fusion for? For somehow emulating what the current fission reactors (logically by bearing in mind the multiple differences between both phenomena) are doing; more specifically, to emulate the stable (and not-too-high-temperature) heat generation outputted by the fission reactions. Why the fission reactors have to generate stable (and not-too-high-temperature) heat? Because all the fission-based power plants (equivalently to what is expected to happen with the future(?) fusion ones) are based on a very simple but perfectly working idea: the electricity is generated from the rotation of a turbine; this turbine has to be regularly rotating for as long periods as possible (because we need all the possible electricity and more); this rotation is achieved by boiling water, phenomenon which happens only its temperature is within certain pretty small range (a bit below/above 100C). The current nuclear (fission) plants get this stable source of heat (with not-too-high-temperature) by taking advantage from what the fission reactions generate. This generated temperature is quite low and that’s why it represents a good source of energy on this context (i.e., a good waiter-boiler). You also don’t need too much energy to start the fission reactions; in fact: you don’t need to start them too often (almost never), this is the beauty of the chain fission reactions and that’s why this form of energy (independently upon the associated dangers) is very efficient and that’s why the nuclear fission plants are the most powerful ones.
- The fusion reactors are expected to apply exactly the same ideas than the fission ones; because the fusion reactions are also expected to generate a regular source of heat (which is basically what we want). The problem is that this time, starting the process (and maintaining it) is not so easy. Additionally the temperature of the generated heat is so high that provokes lots of problems. Creating a chain reaction in fission is very easy; you can see it as a pool (billiard) game: you through a ball which hits another one, which hits the wall, which hits another one After the process is started, it will continue for really long. In fact, without chain reactions nuclear energy would make any sense (= couldn’t considered a form of energy) as far as the heat generated in just one event (fission/fusion) is not too relevant. When you say that fusion hasn’t reached the break-even or that you have put in x and it can only generate x-1, you are not defining a form of energy, but a form of consumption (equivalently to the water which is expected to be boiled).
- Thus, when I talk about fusion I understand a form of energy capable to deliver what is expected, that is: generating heat (at an adequate temperature) for a long enough period of time (to boil water which will then generate electricity). If you are able to fuse two nuclei, excellent; if these first two nuclei generate enough heat to automatically provoke the fusion of the neighbour ones (and so on and so forth), excellent too; if you need lots of energy to perform these fusions but you create a properly working framework, excellent. But don’t call it form of energy i
After a really quick look at the first link, I saw "world-record of 10.7 megawatts" (around 50 times less than a small fission plant?! Impressive! -> just joking; this is a good enough proof to dismiss my point). Anyway... I have to go through all this information to make sure that the name "fusion" is used as I understand it. In any case, this is precisely what I was asking for: thanks.
All what you write after "you are the type of person..." has nothing to do with my personality and what happened here. I seriously thought that “no viable fusion” meant never happened at all (other than in laboratories and under very restricted conditions). This is the reason for my inactivity: not believing that it was possible (-> nothing to do with your interpretation. Curious, isn’t it? Might it be possible that just by asking directly you might understand anything better than by building whole theories based on misinterpretations?). I will take a look and this information and come back to you ASAP.
But why people keep saying things completely against what I am defending without providing any kind of reference? Now high school students can create macroscopic (or even microscopic) fusion reactions!? But where are you getting all these weird ideas from?
Humankind hasn't been able to create fusion reactions with relevant macroscopic effects in their whole history! Currently, many countries are investing lots of money in the ITER project to create the first actually-working fusion reactor ever! They are so lost (because nothing on these lines has ever been done before) that this project is being systematically delayed!
When I firstly read some of the comments to this article, I said to me "wow! Pure knowledge in action!". Why have I to participate in so extremely pointless discussions? How can such worthy comments be written so close to nonsense on the lines of there-are-tons-of-fusion-reactors-because-I-say-so?!
Numerous fusion reactors have been built on different scales.
As requested, please provide reliable references supporting your claims. Just one example with relevant macroscopic effects would be more than enough to me (although well... depending upon the exact conditions the expression "fusion reactor" might not be too accurate).
I'm not sure what you mean by "theoretical".
I meant that they have a very nice theory and are working on implementing it; but with no tangible results so far (other than delays).
Multiple university sized projects already can reach temperatures higher than the center of the Sun. Reaching the temperature is not difficult
By applying these ideas, I might also say: as far as there are many people capable to run 100m in less than 10sec, getting this performance is very easy and thus we can safely assume that lots of people should be able to run 200m in 20sec, 400m in 40sec, etc. That is: the fact of being able to generate certain temperature under very specific conditions (independently upon the number of successful events) does not mean that having this performance is easy or acceptable in any situation (and what is more important: keeping these temperatures for as long as required or starting/stopping the process at will).
Actually, a lot of these problems are much more relevant at smaller scales
?! So... are you saying that the tests are being made under the hardest conditions? Then, why not starting at the macroscopic level right away and creating actually relevant fusion reactions? I do certainly not understand what you are exactly trying to defend/criticise here.
I'm sorry, but when you make statements that look like you haven't even taken the time to look at a Wikipedia level description of fusion power
While studying for my MS in Industrial Engineering specialising in energy, I had quite a few nuclear engineering related subjects. On the other hand, I do understand that many people have a much better knowledge than me on this matter (mainly by bearing in mind that this happened 10 years ago and never worked on it); these people do not even need to have a relevant background on associated issues, like physics or engineering or even basic power-plant knowledge. In any case, the way in which your whole message is written tells me that your knowledge in this specific field/engineering/physics is not too deep; could you please talk a bit about your exact background (are you at the Wikipedia level?).
Sorry but I will not answer any other comment on these lines. I don't want to get involved in this kind of discussions where anything can be claimed based on nothing. My initial request was very clear: please, support all your claims against what I am defending (and what I consider basic understanding of this matter) with reliable references.
Slashdot Mirror
PS: I am not sure about where your reasoning is coming from. What 26.2% at 15 fps means is that 26.2% of the occurrences of the target behaviour (= not necessarily matching a given frame) succeeded. You cannot extrapolate this information to different conditions, like higher/lower number of frames per second.
Good point. Sorry for my short-sighted interpretation.
I took the information from the paper itself, more specifically from the following part:
"For example, when training on the KITTI pedestrian dataset [18], the best known average miss rate is 61.2%, whereas when training on INRIA [10], the average miss rate is 50.2% [6]. Both miss rates are much higher than 31.1% of our method"
As I understand it, a miss rate of 31.1% on a given dataset means that 31.1% of the tested attempts failed.
Idea 1: the fact that accounting for the eventuality of not hitting pedestrians (or any other being/thing) is "one of the unexciting but vital bits of technology" when talking about an autonomous car provides a quite accurate summary about what a big proportion of AI-focused approaches are about. Lots buzz (= exciting technological break-troughs) and not actually-working results (= unexciting technical bits avoiding the big idea to work at all). And this is not just what the OP thinks; Google has been testing autonomous cars for some years already without having still tackled such a secondary(?!) issue.
Idea 2: after quickly skimming through this paper, it seems that the new much-more-accurate algorithm still misses 30% of cases. For me, hurting (even killing) 3 out of 10 pedestrians still sounds quite bad. Additionally, we are talking about their training dataset whose exact complexity is not too clear. For example: what about a kid suddenly crossing the street?, how good is this algorithm at differentiating between persons and similar shapes (human-like advertisement)?, how does it behave in poor-visibility conditions?, etc.
I am completely aware about the tremendous difficulty associated with accomplishing the expected goal and the outputs so far seem promising. But why are they implying that something is almost done, when quite a few basic problems haven't still been tackled?
Sorry to read about your problems, but all this has nothing to do with me: I rarely get angry; and when I do, I would rarely make decisions; and even in case of making decisions, I would always take full responsibility for my actions (I take responsibility for what I did even when being a kid) and eventually re-decide anything.
I am a very sensible guy who has traditionally been surrounded by very sensible people. During the last years, I have been forced to deal with not-too-sensible individuals, the kind of people with whom your statement "when everybody is yelling then nobody is listening" makes sense at all.
As explained in my reply to the comment above, the referred aggressive emails were a gesture of generosity (not of angriness); equivalently to what I have done in quite a few other contexts. But also as explained, I am quite tired now but also sure that, in some situations, I should be more egoist.
This is how I (used to) see things:
1. You are doing something which can be criticised by others without being truly able to defend your position. Example: keep using someone else's email address after a nice request from its rightful owner.
2. You are not fully aware about the aforementioned situation for whatever reason.
3. Someone replies you quite aggressively but by providing lots of useful information.
4 (what I was expecting). Think carefully about what you are exactly doing, understand that you cannot defend your position, learn the lesson and move on.
4 (what I have seen lots of times). People obsessively repeating the same actions by completely altering the original intention. For example: keep sending the same offers to a person who has let very clear his position regarding the product/company.
Among grow-ups I don't have to take care of the education (and/or account for the complexes) of each single person I met. My aforementioned previous (aggressive) attempts were quite generous gestures, which some people might have taken as a help to understand that they were acting wrong.
Your whole speech, although undoubtedly nice and coherent in appearance, actually hides ideas on the lines of "never say you did it wrong", "everyone is right" or similar; the kind of trendy educational attitudes outputting spoiled children with neither boundaries nor respect for others. If you did something wrong, accept the consequences; and stop looking for excuses to somehow compensate your actions.
I have always been an extremely adaptable and understanding person, always thinking that we all are identical and trying to help my fellow men to see their errors. I am very tired now and I have confirmed that generosity can only be delivered within properly-understanding contexts; that’s why I only want to deal with properly understanding individuals. With all the other people, I will rely on the referred “better ignoring them" attitude, much more egoist than my previous behaviour but proven required.
Thanks for having shared so profound thoughts. I do understand the intrinsic limitations of your actions and will overlook the tremendous irony underlying your whole post. The fact that you wanted to ignore what I wrote has indeed hurt my feelings. I hope that other person will inflict you approximately the same pain which I am feeling right now.
PS: I will not say anything about your meta references.
I used to think that ignoring someone's email was the worst you could do to that person. In fact, I used to write lengthy replies with a quite aggressive tone to people doing really stupid things; having ignored them would have been 1000 times easier for me. I wasn't angry at all (I have rarely done anything on the lines of e-venting), but didn't find any other way to communicate with people not able to understand the simplest idea (e.g., spammer continue spamming after you say that are not interested in these products).
The results were really bad. I came to the sad conclusion that there are quite a few people with not much self-respect, who just look for any excuse to blow everything out of proportion and evade all responsibility for their actions (e.g., "why are you screaming to me?", "because you screamed to me before", "I don't care! You cannot scream to me!").
Now I am always using the easiest alternative (although I still think that it is the most aggressive and disrespectful one): I plainly ignore people showing nonsensical behaviours.
Out of two jokes, rightly choosing the funnier one 64% of the times can be considered as almost random; mainly when analysing a pretty small dataset (what is presumably the case here). Bear in mind than reaching the ideal 50% value might be difficult even for a random number generator and lots of attempts might be required.
The most logical way to show results on these lines is by including a better description of the testing conditions (e.g., X samples, by considering Y of them to be negligible because this and that factors and Z to have these peculiarities).
Just as a blind number, 55.8% sounds horrible and 64% sounds still pretty bad. If I was a New Yorker editor, I wouldn't ever trust in such a tool to perform my work. And this is precisely the whole point here: not considering that 64% is good enough because of not having anything better (or because of being a good result for a computer), but because of being actually trustworthy. If you cannot reach such a stage, better keep working and don’t show as a victory what has no real value.
The situation you describe is different to what this algorithm has to deal with (just funny vs. not funny). In any case a 55% success rate would be equally bad in both scenarios.
For example: if out of 10 funny cases, the algorithm concludes that 5 are funny and the other 5 are not; it is irrelevant the fact of having just one not-funny option or as many additional ones as required (interesting, insightful...); the algorithm would fail in a very relevant proportion of the cases. And its failure would be very similar to the one output by a random decision maker (whose theoretical error is precisely 50% when deciding whether something is in some way or not).
Anything below 75% converts almost-understanding into random behaviour in most of the situations. What would you think about a person getting only half of the jokes you tell? Better: about a person just understanding that your sentences are expected to be forming a joke (properly getting the joke is a completely different story).
Using the New Yorker jokes to train the algorithm?!
About 55.8% of the time the humans agree with the captions the AI selects (55.8% of people finding funny the New Yorker jokes, who are less than 10% of all the humans in the world)?!
It seems to me that this is a new generator of almost-random answers with a surprisingly high cost, because their not-particularly-knowledgeable new owners are glad to pay whatever is required to feel that they are part of the AI revolution.
CLARIFICATION: Review (any of the aforementioned versions) is a comedy TV show, a mockumentary, where a guy rates the craziest situations (e.g., starting a cult). It is one of the funniest shows I have ever seen (both versions, although I prefer the American one).
Some people might think that this clarification is not required, mainly after having added a link to a video clearly showing that it is a comedy show. But I have met lots of people lately with serious proper-understanding problems and am quite tired or their noisy nonsense. This is the reason why now I always make completely sure that every single bit is perfectly clear to anyone.
I will think twice (and thrice) before posting a new not-too-direct sarcastic remark here just to avoid this kind of not-funny-at-all clarifications.
Walking on water!? I don't need anything else, It is undoubtedly the new Messiah. This is excellent news for me, as far as I was planning to start a cult anyway.
To confirm how serious I am about all this and to celebrate that the second season of Review (with Forrest MacNeil) started yesterday, here you have a worth-watching video:
https://www.youtube.com/watch?v=jRZIIms7MCU (Starting a cult. Review with Myles Barlow, the Australian prequel of Review).
It was certainly an exclusive. In fact, Mega has defended itself just some minutes ago, by tweeting a reference to this article: https://torrentfreak.com/kim-d...
Being a billionaire nowadays is certainly hard; in fact, this is the main reason why I am poor. There is a tremendous peer pressure because all the other billionaires are donating lots of money to good causes.
I imagine what a billionaire might be thinking while choosing his defining good cause: “This one is not cool enough”; “Bill is already taking care of this other one”; “Is that one a scam? Will keep it on hold for the time being”; “This one is too expensive and spoiling my kids is not precisely cheap”.
Billionaires certainly miss simpler times, when they were keeping all their money to themselves. Poor men/women!
WARNING: this comment contains sarcasm.
Better read firstly my answer to your last message below (my tone is more relaxed).
Sorry, I answered this one also without thinking to much and right after completing my long answer to your comment above. Better read firstly my answer to your last message below (my tone is more relaxed).
Fusion power is defined as the amount of energy per time released by fusion reactions in the experiment,
Nice thing to know. So nothing to do with a standard nuclear (fission) plant, where the power is measured by the electricity being generated. That is: your fusion "reactors" (or test reactors or expensive toys) are measuring the generated power in a completely different way than what is standard (and actually relevant; as far as the whole point of fusion is generating electricity). Is not this a somehow-misleading way to refer to very important concepts? on one hand, you are using the same word (reactor) to refer to two different realities; on another hand, you are measuring different variables but using the same name (generated power).
Besides, your original post said, " macroscopic-relevant fusion reaction," when I think most would consider MW of reactions to be quite macroscopic
Lots of misunderstandings over here, that's why I thought that writing the aforementioned clear-enough summary was required. With macroscopic relevant I mean something which has an actual effect at the macroscopic level, like heating a cup of coffee during 5 minutes. Playing around with atoms is required, but it is just a pre-step. For example, in fission: you have to firstly have control over the fission reactions (start and terminate them at will), then convert the outputted energy in what you want (i.e., source of heat able to keep the temperature in a water tank with certain dimensions at around 100C) and finally make sure that the whole process is stable and regular enough (like a switch you can set on and off to get what you want). With macroscopic effects I meant having a preliminary version (needing whatever crazily big resources are required) of a system capable to perform all the steps until the end, that is: actually generating a usable source of heat. Making sure that the atoms behave as you want is one of the intermediate steps and what I call "microscopic level" as far as my cup of coffee can still not be heated.
A practical fusion reactor does not require a completely self-sustaining chain reaction to run,
Such a statement sounds very compatible with the fusion theory and reality (= labs completely focused on making sure that the first steps work well enough, still very far away from caring about the actual applicability of these first steps), but unfortunately extremely incompatible with the actual reality of electricity generation. As explained in my previous point, what a fusion reactor is expected to do (at least, ITER the first real fusion reactor) is applying the Rankine cycle (the one used in all the fission plants) which expects a regular source of heat at certain temperature. Regular means over months (I think that the standard fission plants are stopped just once or perhaps twice a year just for security reasons; what means that the chain reactions are automatically happening during all that time); you cannot have a power plant generating electricity at irregular intervals. I think that ITER expects to reach more or less self-sustained chain reactions, equivalently to what happens in fission reactors, basically because this is what is required: a regular (over months) source of heat keeping the temperature of the water in a tank within a very narrow range. I am afraid that you cannot get that with "shots" of extremely hot reactions, which can be repeated once every x minutes. Or even in case of getting such a situation (still quite far from that anyway), the resulting system would be extremely inefficient in comparison with any other fuel. Note that additionally to requiring sun-like temperatures to start the process, you have to spend resources on containing the hot plasma (to eventually refrigerate it) and to convert the high temperatures into the target one of around 100C. The only way to make such a crazily inefficient system to be more or less acceptable would be by relying on self-sustained reac
You know about this thing of time zones and time difference, don't you? Is this the kind of thing that a person (well... presumably; an anonymous coward can be many things :) I am joking, don't get angry just for this) with background in plasma physics can know? I am in Spain; the last time I checked this comment was 10pm my time; in any case, why should I come here to answer you immediately?! Why you think that anything you say is relevant? I am systematically defining all what you are saying as nonsense and telling you that "will stop answering you because doesn't make any sense to me"; isn't this enough to understand that your opinion is not important to me? This is kind of funny, because now I am living in a small town in the middle of nowhere, where people show an equivalent behaviour: they seem to think that their opinions have an intrinsic value no matter how many times you say them "what you are saying do not make any sense to me". They are also very nervous and impatient. Anyway...
Funny how you demand references from other people, but act like we should depend on your above average background as you've never provided any references to your claim
What is the exact problem between the reality (= the current conversation which you can read as many times as you wish) and the weird conclusions you draw and the pointless replies you write? At what point have I demand anything?! You were the one doubting about my knowledge and I have told you about my knowledge. I have merely forwarded such a request back. I have ever doubted about anything you said (despite the subtle difference that I am a logged-clearly-identified user and you are discussing as an anonymous coward): you didn’t say plainly anything. You are now doubting about what I am saying?! Seriously, what is wrong with you?
you were given the exact name of a major 15+ year experiment
Have you even gone to the University? For what would an engineer want to know the names of experiments about something of no practical use (not for engineering, not for anyone)? In Industrial Engineering (in Spain, 10 years ago), I got my BS in Mechanical Engineering without hearing a word about nuclear fusion (not even fission). When studying for my MS (4th & 5th), precisely specialising in energy (i.e., power plants and engines) and despite having quite few fission-, power-plant- and electricity-generation-related subjects I didn’t hear a word about fusion, other than side comments on the lines of “never has been done”, “impossible dream”, “during over the last 40 years, many people have promised fusion in 10 years times”, etc. The only time when I studied something related to abstract physics theories was in an introductory subject (which was precisely criticised because of being unnecessarily abstract and useless; BTW I got 9.5 over 10) mentioning Relativity and Quantum Mechanics (BTW perhaps you want to know my opinion about the first one, I call it: “Critical Analysis of the Main Premises of Special Relativity: Lorentz & Minkowski”). Why? Because engineers do not care about these things, because the actual applicability of these theories is none (please, refer to the aforementioned critical analysis)!
I wasn't here talking as an expert in fusion, because I am not (I don't even believe that it is possible). I was talking on account of my knowledge (the one I got in the University and in some later researches because I like the field; but not fusion; nuclear engineering and fission and actually-applicable physics); was sharing what I think that has been done and can be done. I said it very clearly: correct me if I am wrong, but support your claims with reliable references. Why? Because I have a solid enough knowledge to know that certain things haven't happen yet (and what is required to make them happen is so complex and so expensive that are extremely unlikely to happen within the quite a few next years). Or by using other words: I don’t need to be an expert magician to discuss about magic existence. In fact, most of people defending that magic does not exist are not magicians; why would be otherwise? And you are blaming me for not knowing magic? No, I don’t know magic, why should I?
(e.g. JET, EAST, DIII-D, KSTAR, Tore Supra, ASDEX, among many more), plus a couple dozen medium scale projects and other large scale projects of competing design
I talked about this in my comment below (which I think you have also answered; will look at it right after finishing this comment). These are experiments at universities. You can call them reactors because this is the usual name for the container where (nuclear) reactions happen. But you shouldn’t use this term directly (your beloved Wikipedia uses the expression “test reactor”; everything is fine as far as t
I am including below these lines a comment I wrote above as a reply to another person (providing links to various fusion reactors). I think that these ideas are also helpful here to adequately explain my exact position on the fusion-reactors front: ----------------- To avoid future misunderstandings, here you have a set of more detailed (perhaps too evident at some point; but will surely avoid misterpretations) ideas:
- What we want fusion for? For somehow emulating what the current fission reactors (logically by bearing in mind the multiple differences between both phenomena) are doing; more specifically, to emulate the stable (and not-too-high-temperature) heat generation outputted by the fission reactions. Why the fission reactors have to generate stable (and not-too-high-temperature) heat? Because all the fission-based power plants (equivalently to what is expected to happen with the future(?) fusion ones) are based on a very simple but perfectly working idea: the electricity is generated from the rotation of a turbine; this turbine has to be regularly rotating for as long periods as possible (because we need all the possible electricity and more); this rotation is achieved by boiling water, phenomenon which happens only its temperature is within certain pretty small range (a bit below/above 100C). The current nuclear (fission) plants get this stable source of heat (with not-too-high-temperature) by taking advantage from what the fission reactions generate. This generated temperature is quite low and that’s why it represents a good source of energy on this context (i.e., a good waiter-boiler). You also don’t need too much energy to start the fission reactions; in fact: you don’t need to start them too often (almost never), this is the beauty of the chain fission reactions and that’s why this form of energy (independently upon the associated dangers) is very efficient and that’s why the nuclear fission plants are the most powerful ones.
- The fusion reactors are expected to apply exactly the same ideas than the fission ones; because the fusion reactions are also expected to generate a regular source of heat (which is basically what we want). The problem is that this time, starting the process (and maintaining it) is not so easy. Additionally the temperature of the generated heat is so high that provokes lots of problems. Creating a chain reaction in fission is very easy; you can see it as a pool (billiard) game: you through a ball which hits another one, which hits the wall, which hits another one After the process is started, it will continue for really long. In fact, without chain reactions nuclear energy would make any sense (= couldn’t considered a form of energy) as far as the heat generated in just one event (fission/fusion) is not too relevant. When you say that fusion hasn’t reached the break-even or that you have put in x and it can only generate x-1, you are not defining a form of energy, but a form of consumption (equivalently to the water which is expected to be boiled).
- Thus, when I talk about fusion I understand a form of energy capable to deliver what is expected, that is: generating heat (at an adequate temperature) for a long enough period of time (to boil water which will then generate electricity). If you are able to fuse two nuclei, excellent; if these first two nuclei generate enough heat to automatically provoke the fusion of the neighbour ones (and so on and so forth), excellent too; if you need lots of energy to perform these fusions but you create a properly working framework, excellent. But don’t call it form of energy if it does not behave as what is traditionally called form of energy (in this context: stable water boiler). If after years (and years and years) the maximum you get (after having spent imaginably high amounts of money) is some nuclei to fuse, a system able to generate and contain the disproportionately high required temperatures, and even get some chain reactions for some seconds Be very proud of
First of all: thanks for your patience and for making the (small) effort of putting together these ideas. I do recognise that have learned some things which I didn’t know (e.g., the magnetic container proposed by the ITER has already been tested in many different situations, although just during some seconds).
Other than that, I am afraid that my ideas haven't changed appreciably (at least, not in the I-was-wrong direction). I think that there are quite a few not-completely-clear-to-everyone points which provoke systematic misunderstandings. Thus, to reward your effort (, to prove that your interpretation of my behaviour was completely wrong) and to avoid future misunderstandings, here you have a set of more detailed (perhaps too evident at some point; but will surely avoid misterpretations) ideas:
- What we want fusion for? For somehow emulating what the current fission reactors (logically by bearing in mind the multiple differences between both phenomena) are doing; more specifically, to emulate the stable (and not-too-high-temperature) heat generation outputted by the fission reactions. Why the fission reactors have to generate stable (and not-too-high-temperature) heat? Because all the fission-based power plants (equivalently to what is expected to happen with the future(?) fusion ones) are based on a very simple but perfectly working idea: the electricity is generated from the rotation of a turbine; this turbine has to be regularly rotating for as long periods as possible (because we need all the possible electricity and more); this rotation is achieved by boiling water, phenomenon which happens only its temperature is within certain pretty small range (a bit below/above 100C). The current nuclear (fission) plants get this stable source of heat (with not-too-high-temperature) by taking advantage from what the fission reactions generate. This generated temperature is quite low and that’s why it represents a good source of energy on this context (i.e., a good waiter-boiler). You also don’t need too much energy to start the fission reactions; in fact: you don’t need to start them too often (almost never), this is the beauty of the chain fission reactions and that’s why this form of energy (independently upon the associated dangers) is very efficient and that’s why the nuclear fission plants are the most powerful ones.
- The fusion reactors are expected to apply exactly the same ideas than the fission ones; because the fusion reactions are also expected to generate a regular source of heat (which is basically what we want). The problem is that this time, starting the process (and maintaining it) is not so easy. Additionally the temperature of the generated heat is so high that provokes lots of problems. Creating a chain reaction in fission is very easy; you can see it as a pool (billiard) game: you through a ball which hits another one, which hits the wall, which hits another one After the process is started, it will continue for really long. In fact, without chain reactions nuclear energy would make any sense (= couldn’t considered a form of energy) as far as the heat generated in just one event (fission/fusion) is not too relevant. When you say that fusion hasn’t reached the break-even or that you have put in x and it can only generate x-1, you are not defining a form of energy, but a form of consumption (equivalently to the water which is expected to be boiled).
- Thus, when I talk about fusion I understand a form of energy capable to deliver what is expected, that is: generating heat (at an adequate temperature) for a long enough period of time (to boil water which will then generate electricity). If you are able to fuse two nuclei, excellent; if these first two nuclei generate enough heat to automatically provoke the fusion of the neighbour ones (and so on and so forth), excellent too; if you need lots of energy to perform these fusions but you create a properly working framework, excellent. But don’t call it form of energy i
After a really quick look at the first link, I saw "world-record of 10.7 megawatts" (around 50 times less than a small fission plant?! Impressive! -> just joking; this is a good enough proof to dismiss my point). Anyway... I have to go through all this information to make sure that the name "fusion" is used as I understand it. In any case, this is precisely what I was asking for: thanks.
All what you write after "you are the type of person..." has nothing to do with my personality and what happened here. I seriously thought that “no viable fusion” meant never happened at all (other than in laboratories and under very restricted conditions). This is the reason for my inactivity: not believing that it was possible (-> nothing to do with your interpretation. Curious, isn’t it? Might it be possible that just by asking directly you might understand anything better than by building whole theories based on misinterpretations?). I will take a look and this information and come back to you ASAP.
But why people keep saying things completely against what I am defending without providing any kind of reference? Now high school students can create macroscopic (or even microscopic) fusion reactions!? But where are you getting all these weird ideas from?
Humankind hasn't been able to create fusion reactions with relevant macroscopic effects in their whole history! Currently, many countries are investing lots of money in the ITER project to create the first actually-working fusion reactor ever! They are so lost (because nothing on these lines has ever been done before) that this project is being systematically delayed!
When I firstly read some of the comments to this article, I said to me "wow! Pure knowledge in action!". Why have I to participate in so extremely pointless discussions? How can such worthy comments be written so close to nonsense on the lines of there-are-tons-of-fusion-reactors-because-I-say-so?!
Numerous fusion reactors have been built on different scales.
As requested, please provide reliable references supporting your claims. Just one example with relevant macroscopic effects would be more than enough to me (although well... depending upon the exact conditions the expression "fusion reactor" might not be too accurate).
I'm not sure what you mean by "theoretical".
I meant that they have a very nice theory and are working on implementing it; but with no tangible results so far (other than delays).
Multiple university sized projects already can reach temperatures higher than the center of the Sun. Reaching the temperature is not difficult
By applying these ideas, I might also say: as far as there are many people capable to run 100m in less than 10sec, getting this performance is very easy and thus we can safely assume that lots of people should be able to run 200m in 20sec, 400m in 40sec, etc. That is: the fact of being able to generate certain temperature under very specific conditions (independently upon the number of successful events) does not mean that having this performance is easy or acceptable in any situation (and what is more important: keeping these temperatures for as long as required or starting/stopping the process at will).
Actually, a lot of these problems are much more relevant at smaller scales
?! So... are you saying that the tests are being made under the hardest conditions? Then, why not starting at the macroscopic level right away and creating actually relevant fusion reactions? I do certainly not understand what you are exactly trying to defend/criticise here.
I'm sorry, but when you make statements that look like you haven't even taken the time to look at a Wikipedia level description of fusion power
While studying for my MS in Industrial Engineering specialising in energy, I had quite a few nuclear engineering related subjects. On the other hand, I do understand that many people have a much better knowledge than me on this matter (mainly by bearing in mind that this happened 10 years ago and never worked on it); these people do not even need to have a relevant background on associated issues, like physics or engineering or even basic power-plant knowledge. In any case, the way in which your whole message is written tells me that your knowledge in this specific field/engineering/physics is not too deep; could you please talk a bit about your exact background (are you at the Wikipedia level?).
Sorry but I will not answer any other comment on these lines. I don't want to get involved in this kind of discussions where anything can be claimed based on nothing. My initial request was very clear: please, support all your claims against what I am defending (and what I consider basic understanding of this matter) with reliable references.