iiinova — where your digital self keeps evolving

Akshay L Chandra @acl21.bsky.social · 27/08/2026

Did we really get GPT-3 moment in robotics before GTA-6?

Akshay L Chandra @acl21.bsky.social · 14/05/2026

📝 CFP open: #RSS2026 Workshop on Robot World Models 🌐🤖🐨 Submit 4 page novel work or 1 pages abstract of published work! 📅 Deadline: June 5, 2026 More info: 🔗 robot-worldmodels.github.io/submit.html

020

Akshay L Chandra @acl21.bsky.social · 05/05/2026

🚀 Announcing #RSS2026 workshop: Robot World Models 🌐🤖🐨🌐 July 17, 2026 in 🇦🇺! We're bringing together researchers building the next generation of WM for robots — models robots can act in: controllable, long-horizon & grounded in real dynamics. 🔗 robot-worldmodels.github.io

192

Akshay L Chandra @acl21.bsky.social · 09/04/2026

Today I swapped the Franka robot in CALVIN for a human arm and regenerated 6 hrs of "human play" data for free (to quickly iterate over a project idea we have). I don't know what I expected but... it's cute lol 😅

000

Akshay L Chandra @acl21.bsky.social · 25/04/2025

Hyperparameter tuning real-world policy learning by peeking into imagined rollouts inside a world model. Saved us a good few days of iterations. 😅

020

Akshay L Chandra @acl21.bsky.social · 31/03/2025

TIL that dog chew toys make great robot manipulation objects. Colourful, sturdy, durable and full of grip! 🙃

010

Akshay L Chandra @acl21.bsky.social · 24/02/2025

What type of classifier would you use for this data distribution? 🤔🤷‍♂️

100

Akshay L Chandra @acl21.bsky.social · 07/02/2025

In Germany, I'm obviously an Indian. In India, I'm apparently too German.

020

Akshay L Chandra @acl21.bsky.social · 04/02/2025

The owl owns my soul now. Und ich bereue nichts! 🇩🇪

000

Reposted by Akshay L Chandra

gaoyuezhou.bsky.social @gaoyuezhou.bsky.social · 31/01/2025

Can we extend the power of world models beyond just online model-based learning? Absolutely! We believe the true potential of world models lies in enabling agents to reason at test time. Introducing DINO-WM: World Models on Pre-trained Visual Features for Zero-shot Planning.

1208

Akshay L Chandra @acl21.bsky.social · 31/12/2024

Other than CALVIN, are there other tabletop robot manipulation sim environments with a pre-recorded play dataset? 🤔

100

Akshay L Chandra @acl21.bsky.social · 20/12/2024

My IQ temporarily peaked by +200 points just by being here.

050

Akshay L Chandra @acl21.bsky.social · 10/12/2024

Running deep RL on vectorized PyBullet envs that live and render on GPU. It is still not enough to stop our A40 cluster from sweating so much. Maybe I should take the Brax or the PufferLib route... 🤔

120

Reposted by Akshay L Chandra

Pablo Samuel Castro @pcastr.bsky.social · 06/12/2024

i was recently asked to provide 4 "desert island" RL papers. if i were stuck on a desert island i'd hope to have something better to read than #RL papers... but anyway, here's a thread with my choices, maybe you can read them on your flight to @neuripsconf.bsky.social #NeurIPS2024 . Enjoy!

411314

Akshay L Chandra @acl21.bsky.social · 05/12/2024

My CPU is handling my master's thesis better than I am. Respect! #MBRL

010

Reposted by Akshay L Chandra

Eugene Vinitsky 🍒 @eugenevinitsky.bsky.social · 16/11/2024

A physical reasoning agent, trained on millions of problems, that you can test out by coming up with physics puzzles and seeing if it can solve them. It's fun to see if you can stump it: kinetix-env.github.io

kinetix-env.github.io

Investigating large-scale training of RL agents in a vast and diverse space of simulated tasks

4458

Akshay L Chandra @acl21.bsky.social · 15/11/2024

I thought the LED inside the oven was a nice touch. I collected hours of play data with it. Should I now be worried about the Amazon seller no more selling the LED? 🥲 #RL #Robotics

111