Takemoto Koichi

NITA(明石)C 元ロボ研制御
ロボット垢

[Takemoto Koichi @Takemonity]

めっちゃ長生きした

[うめ @umenann]

https://t.co/wJ7AkcBS7D
めちゃくちゃおもろかった

[Takemoto Koichi @Takemonity]

うおおおおおおおおめでとう明石🎉‼️🎉‼️🎉

[Takemoto Koichi @Takemonity]

ロボコン!

[Takemoto Koichi @Takemonity]

サポート材でキモさ倍増してる

[Takumi Kawasetsu, 川節拓実 @takumi_k_jpn]

脳を印刷してって言われたんやけど、キモイ

[Takemoto Koichi @Takemonity]

うおおおおおおおおおお

[Google DeepMind @GoogleDeepMind]

Introducing Gemini 4 Argon – our new frontier model.

It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.

[徹宵 @tesshou]

記憶のフタが開いた
保育園児の頃、爺と喧嘩して家出する！って荷造りしてたら、婆ちゃんが「ひもじくないように」ってアメとかお菓子入れてくれてたんだが、そのうち婆ちゃんが「どこでも困らんように」って３０キロの米をリュックに縛りつけた結果、
担げんくてジタバタするだけになった保育園児が俺

[Takemoto Koichi @Takemonity]

ブラウザとかのテーマを好きなキャラとかの色でできるの面白そう

[Matt DesLauriers @mattdesl]

does Jev understand colour?

incredible model, very cheap and fast. definitely feels like a leap forward for creating new UX/UI paradigms on top of modern language models.

[Takemoto Koichi @Takemonity]

min_{θ, h} (1/2) ‖y - W_L h_{L-1}‖²
s.t. h_i = σ(W_i h_{i-1}), i = 1, ..., L-1
と定式化した時にPCはh_iとσ(W_i h_{i-1})の二乗誤差をペナにしてたけど
よくよく考えたら等式制約なんやからラグランジュの未定乗数法の項入れればいいやん的な話っぽい
確かに。

[Sakana AI @SakanaAILabs]

Introducing PC-ALM, a local-learning alternative to backpropagation.

Our method trains 1000-layer neural nets using only local dynamics, and without backprop.

Blog: https://t.co/bBGCgalqKW

Standard deep learning relies on backpropagation. The brain, however, cannot implement backpropagation, at least not exactly. How can a physical system, such as the brain, solve multilayer credit assignment without explicit use of backprop?

We look for inspiration in two related fields: distributed optimization and NeuroAI.

In NeuroAI, predictive coding asks each neuron activation to solve an energy-based inference problem instead of using a standard forward pass. That inference step can be implemented as energy-minimization dynamics on local prediction errors.

This perspective -- each layer as a dynamical system -- has proven promising, but performance of predictive coding hasn't scaled well with depth. Credit signals at far ends of the network struggle to diffuse into internal layers.

We turn to distributed optimization, generalizing predictive coding to use an augmented Lagrangian instead of energy. This motivation stems back to a classic 1988 paper by LeCun, showing that the Lagrange multipliers of a deep network can be identified with gradients of a supervised loss. The augmented Lagrangian then bridges LeCun's perspective to the standard predictive coding that is used in NeuroAI.

We find that this new perspective yields a natural PC-like alternative to backpropagation, resulting in a method we call PC-ALM. PC-ALM differs from PC in that it introduces dual neurons (Lagrange multipliers) as part of the layer-local dynamics, resulting in each layer acting as a PI feedback control system to minimize local prediction errors.

We find that PC-ALM is capable of propagating signals to seemingly arbitrary depth, especially in deep narrow networks where standard PC struggles to learn.

Ultimately, our motivation here is to understand how distributed physical systems, such as the brain, can compute credit signals using only local coupling and local dynamics.

PC-ALM may also inform deep learning in neuromorphic hardware, where dynamics are cheaper than on GPUs.

Paper: https://t.co/doSZ8mzoyK
Code: https://t.co/rxEDIszVKD

[Takemoto Koichi @Takemonity]

稼働したエージェントをAPI料金に換算したら余裕で賞金超えてそう

[OpenAI @OpenAI]

We’re sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.

The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.

The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.

[Takemoto Koichi @Takemonity]

1万体のエージェントはさすがに延べだと思うけどIMOで金取れるようなのを1万体動かすとか異次元すぎて
他の問題も解けると思わない方が難しい
電気と水ぶっ込めばこれが動くんやからえぐい

[OpenAI @OpenAI]

This model represents a step-function improvement on many benchmarks, and its training is ongoing.

Our internal model group arrived at the Navier–Stokes solution in 88 hours, using around 10,000 coordinating AI agents.

Throughout the effort, we maintained the strict safeguards—including monitoring and isolation—that we apply to all our frontier evaluations.

[Kota Tsuruhara @ove_rl36251]

EVARLでの研究成果が #CoRL2026 に採択されました！
初めての国際学会楽しみです！

[Takemoto Koichi @Takemonity]

さすがにLLMでAGIはできないですやんwwとか思ってたらGPT-6出てきてワンチャンありそう感が増してる

[Takemoto Koichi @Takemonity]

TLに流れてくるAstraのComuter-useの性能えぐすぎて

[Takemoto Koichi @Takemonity]

データセットに気を使ったモデルはいつの時代のAIやねんみたいな画像しか作れんイメージあったけど、生成例��たらガチで競争力あってすごい

[SB Intuitions @sbintuitions]

📢テックブログ更新情報

#SBIntuitions が自社開発した画像生成基盤モデル「Sarashina Image 2.6B」をご紹介✨

モデル構成や学習手法、生成例など詳しく解説しています🔍

詳しくはこちら👇
https://t.co/OZpF4Seema

[Takemoto Koichi @Takemonity]

MetaはLlama4のスキャンダル後 googleよりも息してなかったけどMuseでえぐい巻き返ししてる。

[GIGAZINE(ギガジン) @gigazine]

Metaが「Muse Spark 1.3」を発表、ついにAnthropicやOpenAIの最先端モデルに追いつく性能を達成

[Takemoto Koichi @Takemonity]

これ同音異義語がとかより、できるだけ漢字で書けるところは漢字で書くように学校教育で習慣づけられてるって方が近いと思う。

[Gabriel Mayan Peter(Satoshi Gabu-Chan)🇯🇵🇺🇸🇳🇬 @Mayandeval]

日本語学習アプリを見直していたところ、あることに気づきました。日本の友人たちは、なぜ「たべる」ではなく「食べる」と書くのでしょうか？文字数が変わらないのであれば、ひらがなではなく、あえて複雑な漢字を使うことにはどのようなメリットがあるのでしょうか？

[Takemoto Koichi @Takemonity]

これ参戦ボイスがガキ使のOUTみたいに全部名前言ってから最後に参戦なの良い

[六面サイコロ＠腰痛 @rokumen_saikoro]

47都道府県を覚えられない僕たちのために
バカの白地図アプリを作りました
https://t.co/dd6wPPZYVc

[Takemoto Koichi @Takemonity]

gpt 5.6 luna?早い割にまともに会話できそうなのいいな

[Takemoto Koichi @Takemonity]

改めて見ると北野天満宮のラッシュやばい

[みなみかわ @minamikawaaaa]

あまりに楽しみすぎているsetlog