HackerTrans
トップ新着トレンドコメント過去質問紹介求人

gwf

no profile record

コメント

gwf
·3 か月前·議論
An honest accounting of this history has to call out Sinofsky's role, and it's not pretty. He nearly killed the company in pursuit of power, insisting that everything be brought under his control. Who pushed in 2005 that online and search had to be organized under Windows? That was Sinofsky. He forced the engineering team that most needed to move fast onto three‑year product cycles--sheer insanity, and just one of a 1000 bad calls that ultimately contributed to his ouster from Microsoft. And when it came time to plan his own exit from Microsoft, who did Sinofsky turn to? Jeffrey Epstein, of course, because that's what normal people do. Treat everything Sinofsky writes as an attempt to launder his reputation--then add another layer of skepticism on top of that--because Steven Sinofsky does not live in the same moral universe as the people whose careers and products he derailed.
gwf
·3 か月前·議論
It's the new .NET in that it been so overused as to become almost meaningless.
gwf
·昨年·議論
Not trying to "Schmidhuber" this or anything, but I think my 1999 NIPS paper gives a cleaner derivation and explanation for working on the Jacobian. In it, I derive a Jacobian operator that allows you to compute arbitrary products between the Jacobian and any vector, with complexity that is comparable to standard backprop.

[*] G.W. Flake & B.A. Pearlmutter, "Differentiating Functions of the Jacobian with Respect to the Weights," https://proceedings.neurips.cc/paper_files/paper/1999/file/b...