That’s a hell of a thing to say to a person who has literally been stabbed.

You must work in communications for Trump’s White House.

That’s nuclear grade stupidity.

I would be shocked if there wasn’t a single fuse just for that antenna.

Pretty sure at least half of Adobe’s users are on macOS.

It’s neither here or there, but there’s no “D” in “consequences” either….

<This is your daily complementary pedantic snarky comment>

There is a big difference between fearing you’ll lose your job vs. fearing you’ll lose your house with children possibly inside.

Regardless, these GOP members are fearing violence from the far right side of their party.

I mean, yes, that is bad… but the biennium budget for 2011-2013 was around $60 Billion, which means this project represented about half a percent of the overall budget. The current budget deficit is around $60 Million, but the kicker from 2024 was a record $5.6B in unexpected surplus.

The Oracle debacle is a good representation of why government contracting can be a mess, but I don’t think that project caused such a big long-term financial problem. Putting a toll on the I-5 bridge into/from WA is a massive issue, I agree with you there.

I was at a military industry conference a month or so ago, with several flag level officers giving talks. The total amount of lost years of experience and senior leadership in the military and intelligence agencies is staggering. It is going to take generations to recover.

The same Congress that let Trump unilaterally levy a tax on every American business and citizen?

Ducking is a fairly sacred community thing among Jeep owners, most specifically with Wranglers and now Gladiators. Ducks are always gifted as a random act of kindness, so a Jeep with a lot of ducks suggests that person has been rather active in the community.

To play the devil’s advocate, it’s certainly possible Trump publicly supports the House voting to release the original documents because he either 1) has ironclad assurance that it’ll die in the Senate, or 2) will just pocket veto the bill. There was substantial political downside to them opposing their release, and many many ways for them to kill this whole thing in the future.

Trump doesn’t want the files released, but their Legislative Affairs and Communications Director finally woke up to a strategy that wasn’t political suicide.

Yep, it’s got a DMCA takedown now

This is a monumentally dumb and deadly idea. Want to watch the US cease to exist as a global super power? Good luck maintaining our nuclear arsenal without paying people.

I live in Portland, this image from today pretty well sums up what’s going on here.

The entire thing feels manufactured by design to get people talking about Cracker Barrel.

No, you are correct. Hinton began researching ReLUs in 2010 and his students Alex Krizhevsky and Ilya Sutskever used it to train a much deeper network (AlexNet) to win the 2012 ILSVRC. The reason AlexNet was so groundbreaking was because it brought all of the gradient optimization improvements (SGD with momentum as popularized by Schmidhuber, and dropout), better activation functions (ReLU), a deeper network (8 layers), supervised training on very large datasets (necessary to learn good general-purpose convolutional kernels), and GPU acceleration into a single approach.

NNs, and specifically CNNs, won out because they were able to create more expressive and superior image feature representations over the hand-crafted features of competing algorithms. The proof was in the vastly better performance, it was a major jump when the performance on the ILSVRC was becoming saturated. Nobody was making nearly +10% improvements on that challenge back then, it blew everybody out of the water and made NNs and deep learning impossible to ignore.

Edit: to accentuate the point about datasets and GPUs, the original AlexNet developers really struggled to train their model on the GPUs available at the time. The model was too big and they had to split it across two GPUs to make it work. They were some of the first researchers to train large CNNs with GPUs. Without large datasets like the ILSVRC they would not have been able to train good deep hierarchical convolutions, and without better GPUs they wouldn’t have been able to make AlexNet sufficiently large or deep. Training AlexNet on CPU only for ILSVRC was out of the question, it would have taken months of full-tilt, nonstop compute for a single training run. It was more than these two things, as detailed above, but removing those two barriers really allowed CNNs and deep learning to take off. Much of the underlying NN and optimization theory had been around for decades.

Before AlexNet, SVMs were the best algorithms around. LeNet was the only comparable success case for NNs back then, and it was largely seen as exclusively limited to MNIST digits because deep networks were too hard to train. People used HOG+SVM, SIFT, SURF, ORB, older Haar / Viola-Jones features, template matching, random forests, Hough Transforms, sliding windows, deformable parts models… so many techniques that were made obsolete once the first deep networks became viable.

The problem is your schooling was correct at the time, but the march of research progress eventually saw 1) the creation of large, million-scale supervised datasets (ImageNet) and 2) larger / faster GPUs with more on-card memory.

It was fact back in ~2010 that SVMs were superior to NNs in nearly every aspect.

Source: started a PhD on computer vision in 2012

As somebody who has been in one of those data centers for work, good luck getting anywhere near an actual server.

Oregon has been entirely state-wide mail-in voting for decades. Good luck shutting down an entire state’s mail for more than a month.