Thomas Edlich
@grumpybavarian
Science at Google DeepMind. AlphaQubit. Ex-BenevolentAI. Opinions are my own.
🧵 Check out our latest preprint: "Programming by Backprop". What if LLMs could internalize algorithms just by reading code, with no input-output examples? This could reshape how we train models to reason algorithmically. Let's dive into our findings 👇
Introducing Gemini 2.0, our most capable AI model yet designed for the agentic era. Gemini 2.0 brings enhanced performance, more multimodality, and new native tool use.
Really excited to share our (@GoogleDeepMind and @GoogleQuantumAI) new publication in @Nature, where we use neural networks to identify errors in quantum computers with state-of-the-art accuracy. Paper: nature.com/articles/s4158… Blog: blog.google/technology/goo…
I’m so proud of the updated version of #MusicFXDJ we developed in collaboration with @jacobcollier, available today at labs.google/musicfx. Over the past year I’ve spent countless hours experimenting with our real-time music models, and it feels like I’ve learned to play a…
It’s here! Thrilled to collab with @jacobcollier on our latest #LabSession exploring the magical possibilities of generative music and #MusicFXDJ. Watch full video: youtu.be/y7gKlzvg8xk Tune in 10/24 at 5 PM ET as Jacob livestreams a MusicFX DJ sesh on his YT channel.
Check out our paper on how an agent can learn to unknot through reinforcement learning! arxiv.org/abs/2409.09032
Another example of how reinforcement learning can help with exploring the space of mathematical conjectures - both in the broader sense, but also in more narrowly defined subdomains, for more specific questions:
We’re presenting the first AI to solve International Mathematical Olympiad problems at a silver medalist level.🥈 It combines AlphaProof, a new breakthrough model for formal reasoning, and AlphaGeometry 2, an improved version of our previous system. 🧵 dpmd.ai/imo-silver
1 / 🧵 Excited to introduce our #ICML2024 paper: 😈 EvIL (Evolution Strategies for Generalisable Imitation Learning) a new inverse RL (IRL) method for sample efficient transfer of expert behaviour across environments – it's so good, it's downright EvIL!
Welcome to the next chapter in the Gemini era. ✨ We’re bringing our models to more @Google products you know, use & love. From today, Bard will be known as Gemini — and you can access our most capable model, Ultra 1.0, with Gemini Advanced. → dpmd.ai/bard-gemini-ad… #GeminiAI
1/🧵Jax fans, buckle up! Prepare to ditch slowpoke IRL training with the ⚡️jaxirl⚡️ launch, a JAX library built for Inverse Reinforcement Learning algorithms that's 250x faster 🏎️ 💨 than PyTorch. Read on to learn more ⬇️⬇️⬇️
New Research Engineer opening in the DeepMind science team: boards.greenhouse.io/deepmind/jobs/… - please share with your networks!
A Neural Network Solves and Generates Mathematics Problems by Program Synthesis: Calculus, Differential Equations, Linear Algebra, and More abs: arxiv.org/abs/2112.15594 first work to automatically solve, grade, and generate university-level Mathematics course questions at scale
Can machine learning explain the world at the quantum level? New in @ScienceMagazine, with the code also released, our team shows that neural networks can radically improve density functionals with #DM21. Read more: dpmd.ai/density-functi… Code: dpmd.ai/density-functi… 1/
A new paper in @Nature details how machine learning was used to make significant new discoveries in pure mathematics by guiding the intuition of some of the world’s top mathematicians from @Sydney_Uni and @OxUniMaths: dpmd.ai/maths-blog Paper: dpmd.ai/nature-maths 1/
Many single protein chains can be accurately predicted using AlphaFold, but multi-chain protein complexes remain challenging. Introducing AlphaFold-Multimer, a model that can predict the structure of multi-chain protein complexes with increased accuracy dpmd.ai/alphafold_mult… 1/
Published today in @naturemethods together with colleagues from @calico: Enformer - a transformer model that has led to greatly increased accuracy in predicting gene expression from DNA sequence. Blog: dpmd.ai/enformer Paper: dpmd.ai/naturemethods_… 1/
From packing an umbrella to preparing for extreme conditions, predicting short term weather patterns is crucial for daily life. New research with the @metoffice and SOTA model advances the science of Precipitation Nowcasting - the prediction of rain: dpmd.ai/nowcasting 1/4
Great new paper from @DrJoshuaBox & @benedekfabian on molecular design methods. doi.org/10.1016/j.drud… #CompChem #Cheminformatics #Chemoinformatics #ArtificialIntelligence #MachineLearning
Our (long overdue) little pcm paper is on chemrxiv 😀💙: Proteochemometric Models Using Multiple Sequence Alignments and a Subword Segmented Masked Language Model tinyurl.com/2734cycu