Donal Byrne
@DonalByrne2
Followers
164
Following
370
Media
4
Statuses
106
Senior Research Engineer @instadeepai, excited about building large scale RL, optimistic that the robots will remember me fondly
London, England
Joined November 2011
RT @AlexLaterre: Just weeks after releasing ProtBFN/AbBFN in @NatureComms, we're back with AbBFN2 ⚡️. Our new Antibody Foundation Model goe….
0
9
0
RT @tomdbarrett: Excited to announce our latest work: AbBFN2, a generative antibody model co-modelling sequences, genetic origins, and deve….
0
1
0
RT @tomdbarrett: Excited to share our latest work applying Bayesian Flow Networks (BFNs) to proteomics!. We show how BFNs can outperform le….
0
11
0
RT @instadeepai: Excited to share our latest work on Sequential Monte Carlo Policy Optimisation (SPO)🔥— a scalable, search-based RL algorit….
0
7
0
RT @AlexLaterre: ✨ Today in Nature Methods (@naturemethods), we're presenting Nucleotide Transformers (NT), our landmark Foundation Models….
0
1
0
RT @ClementBonnet16: Introducing Latent Program Network (LPN), a new architecture for inductive program synthesis that builds in test-time….
0
31
0
The next addition to my progress in learning C 🤓 is a very simple MLP on MNIST in a little under 500 lines (would be less without data prep and logging), with a comparison speed benchmark in pure Python. Next step is writing it in CUDA 🚀! .
2
2
12
If you are doing exciting work in ML and would like to share it with the community, please feel free to reach out and join one of our future sessions!.
0
0
4
Huge thanks to all of our amazing speakers who have contributed to our series so far ❤️.@vwxyzjn, @MatteoBettini98, Paul Daoudi, @The_Eimer, @jo_brandstetter, @CULLYAntoine.
1
0
4
Very excited to announce @instadeepai's #LetsTalkResearch series where we invite top researchers in ML to share their latest work 🧑🔬. Check out the first series of speakers on our youtube channel #AI #machine_learning #rl #GenAI.
1
4
11
RT @instadeepai: 🚀 We’re proud to announce Kyber, our near-exascale supercomputer, built to drive the next generation of AI research!. Powe….
0
10
0
RT @hannah_janmo: I’m excited to share work that I’ll be presenting at @GeccoConf next month, where we use Quality-Diversity Algorithms for….
0
8
0
We had a great time meeting lots of people yesterday during our poster session on Jumanji at #ICLR2024 . Incase you missed it here is a very brief overview of the work (apologies for the background noise 🙈).
Got lost in the #ICLR2024 poster maze? Don't worry, we've got your covered! 🛟. Here is @DonalByrne2, Senior Research Engineer at.@instadeepai , as he showcases Jumanji — our library for high-performance RL environments in #JAX. ⭐️ Github:
1
2
20
Join me and my co-authors @ClementBonnet16, Daniel Luo and @AlexLaterre tomorrow at 10:45 to discuss Jumanji, our open-source work on high performance #RL environments in #JAX.
Presenting Jumanji at #ICLR2024, our open-source suite of RL environments, designed to be fast, flexible, and scalable, all written in JAX! (1/3). Full paper: Github:
0
7
26
InstaDeep are at #ICLR2024 this week. Please reach out to me if you would like to chat about #JAX and Large Scale ML!.
All eyes on Vienna this week as #ICLR2024 is about to kick off! 🎵. We're thrilled to contribute this year with six papers sharing our team’s work in Decision-Making AI and our latest research on Machine Learning for Biology 🧬. Read more →
0
2
15
@_ddjohnson Turns out there is already good support for working with and converting Flax models 😁
1
0
3
This looks very cool! Would love to see a wrapper for converting Flax models to Penzai for post-training analysis @_ddjohnson 👀.
Excited to share Penzai, a JAX research toolkit from @GoogleDeepMind for building, editing, and visualizing neural networks! Penzai makes it easy to see model internals and lets you inject custom logic anywhere. Check it out on GitHub:
1
1
5
RT @ChalumeauFelix: [1/n] QDax is now published in the JMLR Open Source Software track. Hope this will give a broader exposition to the pa….
0
8
0