Aditya Mohan

PhD Student working on Generlization and State abstractions in #RL, #MetaLearning, and #AutoRL

Previously: @AmazonScience @awscloud

[Aditya Mohan @amsks96]

How can we make Reinforcement Learning (#RL) more deloyable? Can we develop a unified lens for different approaches to tackling this?

Excited to share our survey on Structure in RL, with @yayitsamyzhang and @LindauerMarius

https://t.co/GXOOa7877x
(1/7)

[Aditya Mohan @amsks96]

@harshit_sikchi I just read SF as Successor Features 🤣

Seems like a good time to take a break

[Aditya Mohan @amsks96]

@twni2016 @RBCBorealis @pierrelux Congrats!!

[Aditya Mohan @amsks96]

@c_voelcker That might make the question much harder :P

[vmoens @VincentMoens]

Stoked to announce that BricksRL, our paper on using PyTorch/TorchRL to train the LEGO® robots in the wild has been accepted at @NeurIPSConf as a spotlight paper! 🪢

[Pablo Samuel Castro @pcastr]

Final #runconference @RL_Conference (with my brother as special guest)!!
This has been an amazing set of runs, and I think it's the first time we've had so many people attend all of them.
Thanks to everyone who came out!
��🏾⛹🏾💨🤖

[Aditya Mohan @amsks96]

Interested in ways of exploiting problem structure in RL? Come to our poster at @RL_Conference today!

https://t.co/IbIGZFhZfP

[Aditya Mohan @amsks96]

How can we make Reinforcement Learning (#RL) more deloyable? Can we develop a unified lens for different approaches to tackling this?

Excited to share our survey on Structure in RL, with @yayitsamyzhang and @LindauerMarius

https://t.co/GXOOa7877x
(1/7)

[Aditya Mohan @amsks96]

Super cool talk by Prof. Doshi-Velez at @RL_Conference highlighting the importance of utilising problem structure when deploying RL on real-world settings!

[Pablo Samuel Castro @pcastr]

Third keynote by Andy Barto @RL_Conference , arguing that it was always RL, with a standing ovation at the end!

[Aditya Mohan @amsks96]

Excellent talk by Doina Precup at @RL_Conference on rethinking the foundations and the necessity of Structure in RL!

[Pablo Samuel Castro @pcastr]

Great turn out for the first #runconference @RL_Conference !

[Aditya Mohan @amsks96]

Super excited to attend @RL_Conference in a few days and can’t wait for the amazing discussions!

If anyone needs an overview of all the papers (Main Conf, Journals, Workshops), here's a scraped Excel Sheet: https://t.co/s0SUusFsVK

[Aditya Mohan @amsks96]

@jparkerholder Hope you all feel better soon

[Aditya Mohan @amsks96]

@CarolaDoerr19 @ERC_Research @LIP6_lab @CNRS @CNRSinformatics @Sorbonne_Univ_ Heartiest congratulations!! Looking forward to a lot more research in this area 😁

[Aditya Mohan @amsks96]

We all love cakes by @The_Eimer

[Aditya Mohan @amsks96]

@pcastr @jacobcollier Oh yes, Naturally very innovative and different form the original version! Loved the vibe

Summer rain was also a feast!

[Aditya Mohan @amsks96]

Things just keep getting better for JAX-Based environments!!

[Michael Matthews @mitrma]

I’m excited to announce Craftax, a new benchmark for open-ended RL!

⚔��� Extends the popular Crafter benchmark with Nethack-like dungeons
⚡Implemented entirely in Jax, achieving speedups of over 100x
1/

[Aditya Mohan @amsks96]

@AvivTamar1 @wielandbr Theorem 2 was pretty neat! Enjoyed reading the paper

[Greg Brockman @gdb]

Announcing Sora — our model which creates minute-long videos from a text prompt: https://t.co/SZ3OxPnxwz

[Aditya Mohan @amsks96]

Nice overview of the work on Successor Representations, Features, and related ideas!

[Wilka Carvalho @cogscikid]

Excited to share our review on the successor representation and two important generalizations: successor features and successor models!

These provide answers spanning AI, cognitive science, and neuroscience for how agents can efficiently reuse their knowledge in new situations