Research Scientist with Parsons working at ARL. I work on (MA)RL and other ML things.
🎸@lostcontinentde
Opinions are all mine
@[email protected]Joined February 2020
Introducing Project Sid: the first simulations of 1000+ truly autonomous agents collaborating in a virtual world, w/ emergent economy, culture, religion, and government
Humans are the only species to land the moon, because we can cooperate at a vast scale
Can AI do the same?
Can you solve MARL tasks without conditioning the policy on the observation? In our new paper we investigate the popular StarCraft micromanagement benchmark SMAC and discover that most scenarios do not require the observation to be used! [1/N]
💯 Been saying this since my first neuromorphic project in 2017 – someday we're gonna see a neuromorphic-like chip become standard in PCs like the GPU has
💯 Been saying this since my first neuromorphic project in 2017 – someday we're gonna see a neuromorphic-like chip become standard in PCs like the GPU has
Cost for one vial of insulin:
🇦🇺: $6.94
🇬🇧: $7.52
🇫🇷: $9.08
🇩🇪: $11
🇨🇦: $12
🇯🇵: $14.40
🇺🇸: $98.70
No one should be forced to take out their wallets for insulin – a life-saving drug that was invented nearly 100 years ago. It's time to end the greed of the pharmaceutical industry.
Poster on the taxonomy of main statistical distances with their underlying geometries: Euclidean geometry, Riemannian & Finslerian geometry, information geometry, affine differential geometry, quantum geometry, etc.
- PDF: tinyurl.com/TaxonomyDistan…
- Portal: franknielsen.github.io/Divergence/ind…
Using HyperTree Proof Search we created a new neural theorem solver that was able to solve 5x more International Math Olympiad problems than any previous AI system & best previous state-of-the-art systems on miniF2F & Metamath.
More in our new post ⬇️
1. This thread is about the report of the UN Independent International Commission about the russian atrocities and violence in Ukraine.
It is hell.
I focus here only on the cases of rape and violence.
Plese STOP russia.
In our new work - Algorithm Distillation - we show that transformers can improve themselves autonomously through trial and error without ever updating their weights.
No prompting, no finetuning. A single transformer collects its own data and maximizes rewards on new tasks.
1/N
Today we're launching the Farama Foundation, a new nonprofit dedicated to open source reinforcement learning, and we're beginning by maintaining and standardizing all the major open source reinforcement learning environments. Read more here: farama.org/Announcing-The…
Today, we’re introducing a new model that eliminates accept/reject decisions.
By publishing every paper with eLife reviews as a Reviewed Preprint, we plan to restore autonomy to authors, ensuring that they will be judged by what, not where, they publish. elifesciences.org/inside-elife/5…
After building on years of work from MILA, DeepMind, ourselves, and others, our AIs are now expert-human-level in no-press Diplomacy and Hanabi! Unlike Go and Dota, Diplomacy/Hanabi involve *cooperation*, which breaks naive RL. arxiv.org/abs/2210.05492arxiv.org/abs/2210.05125 🧵👇
"A Year"
AI animation artwork made in colab using my custom stable 3D animation algorithm on top of #stablediffusion model. In the thread I share some details about the algo and when i plan to release it, and talk about the joy and future of AI filmmaking
🎶 DakhaBrakha - Vesna
1/7 We built a new model! It’s called Action Transformer (ACT-1) and we taught it to use a bunch of software tools. In this first video, the user simply types a high-level request and ACT-1 does the rest. Read on to see more examples ⬇️
786 Followers 581 FollowingAsk me about #ReinforcementLearning
#AI research @SonyAI_global
RL for games, robotics, and other real-world applications
Views and tweets are my own.
79K Followers 4K FollowingConnect, Unify, Respond to any data, anywhere with human-level intelligence.
@Forbes Most Promising AI Companies, @Gartner_inc Cool Vendor
635K Followers 398 FollowingFree Stock Market News that is FAST, ACCURATE, CONSISTENT, and RELIABLE | Not Just Stock News | My Daily Stock Market Recap is the link in my bio ⬇️
5K Followers 518 FollowingMachine Learner by day, Statistician at ❤️ In search of statistical intuition for modern ML &simple explanations for complex things 👀 (Opinions my own)
786 Followers 581 FollowingAsk me about #ReinforcementLearning
#AI research @SonyAI_global
RL for games, robotics, and other real-world applications
Views and tweets are my own.
22K Followers 1K FollowingDEVCOM Army Research Laboratory is the Army’s research laboratory. ARL’s mission is to operationalize science.
contact: [email protected]
7K Followers 199 FollowingInstaDeep offers a host of Enterprise AI products, ranging from GPU-accelerated insights to self-learning decision making systems.
79K Followers 4K FollowingConnect, Unify, Respond to any data, anywhere with human-level intelligence.
@Forbes Most Promising AI Companies, @Gartner_inc Cool Vendor
242K Followers 21 FollowingWe’ll help you make it like nobody’s business. Multimodal media generation and editing tools to get your idea to production. Self-deploy? 👍 Need a partner? 🤝
842 Followers 2 FollowingHow to ML is a series of non-technical ML research advice posts by @j_foerst and @_rockt.
profile picture by Stable Diffusion
banner by charlesdeluvio
15K Followers 168 FollowingResearch scientist @GoogleDeepMind. Past: @Databricks, first hire @MosaicML, @MIT PhD. I post about AI technical progress + sometimes the business side.
3K Followers 1K Followingwriter/comedian • ucb performer • doesn’t need hateration in the dancery, although holleration may be acceptable on a case by case basis • he/him • bi 🏳️🌈
7K Followers 15 FollowingWe're rekindling the dream of personal computing by making reliable software creation accessible to all. Join us: https://t.co/9UF0rR6YUr
3K Followers 95 FollowingTalkRL Podcast is All Reinforcement Learning, All the Time.
Follow for interviews with brilliant folks from across the world of RL.
Host @robinc. DMs open.
50K Followers 3K FollowingDeveloper Experience Lead at @GoogleDeepMind
Building Gemini API, Gemma, AI Studio and more AI products. My views
ex-Chief Llama Officer @huggingface 🇵🇪🇲🇽
636K Followers 35 FollowingWe're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant @claudeai on https://t.co/FhDI3KQh0n.