Only 3% of energy in Wa state is carbon based (1.2% coal, 1.6% natural gas, all other non-carbon based, hydro, nuclear, wind, etc.). They don't need to get from 85% to 100%, they need to get from 97% to 100%.
I think there is a difference between the Foxconn and Amazon deals. Amazon was really going to bring jobs to the area (see Seattle for evidence).
You may not agree with the trade off between the benefit of the jobs and any tax breaks agreed to (I'm not sure if it was a good, bad or neutral deal for NYC), but there definitely were going to be a significant number of jobs.
to be able to build the floating cities on Mars. Once we get to mars and terraform to create bodies of water, we are going to need these floating cities, so we better figure it out now.
GAN's are a nice technique but the major breakthrough in NN's was the development of learning algorithms for deep networks. Backprop doesn't work due to increasing impact of changes as you work backward through the NN layers.
Hinton and company created a method that gets around that major issue, and it is that method that led to the explosion of successful application of NN's to various problems in the last 10 years.
For example, prior to the deep learning algorithm, the best image recognition was from an evolved multi-layer NN. But evolving NN's is slow, inefficient and not guaranteed to arrive at a reasonably good solution. With Hinton and team's method, the floodgates opened and it was possible to train the network to achieve the best results.
Most of the study used much higher power, but one portion used 1.5 W/kg RFR which is less than the FCC limit, and they found increased rate of cancer in that portion also.
In addition, the US Air Force study from 1990 on radar with pretty low intensity RFR/microwaves (2,450 MHz pulsed radiation with SAR 0.15-0.4 W/kg) found increased incidence of cancer compared to the control group.
Some details from Air Force study:
"primary cancers were detected in the exposed rats 3.6 times more often than in control animals. Additionally, metastatic cancers were detected in the exposed rats 2 times more often than in the control."
How does a video of their object detection system indicate they can handle city driving? You do understand that object detection is just one component of being able to figure out what to do, right?
I know, I was laughing as I read that ("can handle city driving") and watched the video. Just detecting objects is not enough for self-driving, you also need to determine what actions to take. Nothing in that video indicates they've figured out the second part.
Too true! I also have no sympathy for people that complain about cpu performance,if you're a cheapskate and just buy Intel, then you get what you paid for. If you put in some time and resources, you can create an ASIC specific for your workload that will outperform pretty much anything.
"cold fusion": while it may not be fusion, research by US Navy and companies like Mtsubishi Heavy Industries and Toyota is continuing to get more and more consistent excess heat results. Current hypothesis is conversion of protons and electrons into neutrons.
Last years paper from Japan: https://www.sciencedirect.com/science/article/pii/S0360319918320925,
US Navy essay: https://www.usni.org/magazines/proceedings/2018-09/not-cold-fusion
Thank you, excellent info. I went to the ITEP website and to fact check web sites to find out if the story was really telling the entire and accurate story. I found that, while ITEP tends to be highly factual, they are not unbiased in their presentation of the facts.
You just confirmed that bias and at the same time filled in the additional piece of the puzzle showing that taxes were paid but by a different party. Thanks again.
Slashdot Mirror
Let's re-animate his brain to help solve the murder mystery
It targets people that steal.
Are you saying that Amazon should not deduct business expenses like other companies? Or should they only not deduct specific business expenses?
Only 3% of energy in Wa state is carbon based (1.2% coal, 1.6% natural gas, all other non-carbon based, hydro, nuclear, wind, etc.). They don't need to get from 85% to 100%, they need to get from 97% to 100%.
I laughed at both comments :)
I think there is a difference between the Foxconn and Amazon deals. Amazon was really going to bring jobs to the area (see Seattle for evidence).
You may not agree with the trade off between the benefit of the jobs and any tax breaks agreed to (I'm not sure if it was a good, bad or neutral deal for NYC), but there definitely were going to be a significant number of jobs.
to be able to build the floating cities on Mars. Once we get to mars and terraform to create bodies of water, we are going to need these floating cities, so we better figure it out now.
This is project management 101.
Or you will buy a sturdy-ish reusable mat with a preprinted QR code. You scan your landing pad when placing order.
GAN's are a nice technique but the major breakthrough in NN's was the development of learning algorithms for deep networks. Backprop doesn't work due to increasing impact of changes as you work backward through the NN layers.
Hinton and company created a method that gets around that major issue, and it is that method that led to the explosion of successful application of NN's to various problems in the last 10 years.
For example, prior to the deep learning algorithm, the best image recognition was from an evolved multi-layer NN. But evolving NN's is slow, inefficient and not guaranteed to arrive at a reasonably good solution. With Hinton and team's method, the floodgates opened and it was possible to train the network to achieve the best results.
Is it really the same story? or is it a unique instance of this story?
Announcement: "Will the passenger with the loaded gun magazine that already made it past security please return to security, thank you"
It also probably had nothing to do with SQL, 25 txns per second is barely a pulse.
Python is a strongly typed language, so yes, it understands types just like Java. Maybe you are really referring to dynamic vs static?
Great post but I'm still left wondering, how many members? And how much did they pay for extra features?
Should be easier using the new Stanford AI Photo Fake (TM) system.
Sooooooo......these bits go to eleven?
Most of the study used much higher power, but one portion used 1.5 W/kg RFR which is less than the FCC limit, and they found increased rate of cancer in that portion also.
In addition, the US Air Force study from 1990 on radar with pretty low intensity RFR/microwaves (2,450 MHz pulsed radiation with SAR 0.15-0.4 W/kg) found increased incidence of cancer compared to the control group.
Some details from Air Force study: "primary cancers were detected in the exposed rats 3.6 times more often than in control animals. Additionally, metastatic cancers were detected in the exposed rats 2 times more often than in the control."
I'm going to make my new coin AssCoin it will be for when you need to pay but don't like the service you received.
For me and apparently many in Wa state, the best part of DST is sunlight in the evening, after work etc. Lamps are unrelated to that argument.
How does a video of their object detection system indicate they can handle city driving? You do understand that object detection is just one component of being able to figure out what to do, right?
I know, I was laughing as I read that ("can handle city driving") and watched the video. Just detecting objects is not enough for self-driving, you also need to determine what actions to take. Nothing in that video indicates they've figured out the second part.
Too true! I also have no sympathy for people that complain about cpu performance,if you're a cheapskate and just buy Intel, then you get what you paid for. If you put in some time and resources, you can create an ASIC specific for your workload that will outperform pretty much anything.
"cold fusion": while it may not be fusion, research by US Navy and companies like Mtsubishi Heavy Industries and Toyota is continuing to get more and more consistent excess heat results. Current hypothesis is conversion of protons and electrons into neutrons.
Last years paper from Japan: https://www.sciencedirect.com/science/article/pii/S0360319918320925,
US Navy essay: https://www.usni.org/magazines/proceedings/2018-09/not-cold-fusion
And 90% of that profit is from AWS, not from selling merchandise.
Thank you, excellent info. I went to the ITEP website and to fact check web sites to find out if the story was really telling the entire and accurate story. I found that, while ITEP tends to be highly factual, they are not unbiased in their presentation of the facts.
You just confirmed that bias and at the same time filled in the additional piece of the puzzle showing that taxes were paid but by a different party. Thanks again.