Harshit Sikchi
@harshit_sikchi
Followers
2K
Following
2K
Media
43
Statuses
416
Research at @OpenAI; Reinforcement Learning; PhD from UT Austin. Previously FAIR Paris @AIatMeta, @CMU_Robotics @NVIDIAAI @UberATG.
San Francisco, CA
Joined July 2018
Check out GPT-5. Starting around two months ago now, was fortunate to get to contribute to something so fun!.
0
0
13
RT @SebastienBubeck: Claim: gpt-5-pro can prove new interesting mathematics. Proof: I took a convex optimization paper with a clean open p….
0
1K
0
It has been a good conference @RL_Conference ; Below @RLBRew_RLC social, edmonton flame, a great talk. Conference detox needed now
1
1
32
RT @pcastr: Super thought-provoking talk by Dale Schuurmans @RL_Conference on LLMs and computation, and why value-based RL doesn't (or can'….
0
30
0
RT @gdb: Just released gpt-oss: state-of-the-art open-weight language models that deliver strong real-world performance. Runs locally on a….
0
523
0
3. Implicit vs. Explicit Offline Inverse Reinforcement Learning: A Credit Assignment Perspective.Link: We mechanisticallt investigate the differences in implicit and explicit IRL.
openreview.net
Inverse reinforcement learning (IRL) alleviates the practical challenges of reward design by extracting reward functions from approximately rational demonstrators. Despite enjoying theoretical...
0
0
1
2. Regularized Latent Dynamics Prediction is a Strong Baseline For Behavioral Foundation Models. We show that simply using regularized latent dynamics prediction leads to a strong baseline for behavior foundation models; sidestepping a lot of complexities.
openreview.net
Behavioral Foundation Models (BFMs) have seen some success recently in producing agents with the capabilities to adapt to any unknown reward or task. In reality, these methods are only able to...
2
0
0
1. A Unified Framework for Unsupervised Reinforcement Learning Algorithms.Link: We present a unifying perspective on how to connect a long line of research in unsupervised RL.
openreview.net
Many sequential decision-making domains, from robotics to language agents, are naturally multi-task on the same set of underlying dynamics. Rather than learning each task separately, unsupervised...
1
0
0
At @RLBRew_RLC today we are presenting 2 works on unsupervised RL and 1 work on inverse RL. Stop by the poster session to learn more! Details below:.
1
1
8
RT @gio_ramponi: Giving two talks tomorrow at @RL_Conference on Imitation Learning and IRL in multi-agent systems.See you at 11am at @RLBRe….
0
4
0
I will be @RL_Conference presenting the below work on Fast Adaptation on Wednesday August 6 at 10:20 am and some works on unsupervised RL and imitation at RLBrew workshop on August 5.
Behavioral Foundation Models (BFMs) trained with RL are secretly more powerful than we think. BFM’s directly output a policy believed to be near-optimal given any reward function. Our new work shows that they can actually do much better:
3
10
107
The social follows our workshop on RL Beyond Rewards: Check out our schedule here:
0
1
3
RSVP on the link below as the venue has limited space. We saw a big attendance last year!.
lu.ma
Come join us to discuss RL ideas and meet people over food and drinks (sadly not sponsored), find collaborators and friends!
1
1
3
We are hosting a social again this year at #RLC2025 (@RL_Conference ) on August 5. Come to meet people pre-conference and find friends and collaborators. RSVP below if you can make it:
1
8
39
Exploration is crucial for the next breakthrough in reasoning; .Behavioral Foundation Models may change the way we do low level control;.Come to discuss about all these topics and more @RL_Conference !.
🗓️ Mark your calendars to join us in Edmonton on Aug 5 for: . Reinforcement Learning Beyond Rewards: Ingredients for Generalist Agents . With 🔥 speakers:.@chelseabfinn , George Konidaris, @furongh,.@gio_ramponi, @MichaelD1729, Ahmed Touati. #RLC2025 #RL @RL_Conference
0
8
33
Seeing A.R. Rahman in the office: cool perks of the job :D.
It was a pleasure meet @sama at his office …we discussed “Secret Mountain”, our virtual global band, and to empower and uplift Indian minds to use AI tools to address generational challenges and lead the way forward. EPI.@chatgptindia @OpenAI #arrimmersiveentertainment
2
0
20