-
University of California, Berkeley
- New York City Metropolitan Area
- linkedin.com/in/arvind-rajaraman
Highlights
- Pro
Pinned Loading
-
language-skills-diayn
language-skills-diayn PublicDLSD (Diverse Language-based Skill Discovery) is an unsupervised RL method that uses language embeddings as a semantic prior to guide skill discovery. We see that in long-horizon tasks (e.g. the Cr…
Jupyter Notebook 1
-
closed-loop-influence
closed-loop-influence PublicHumans have internal representations of the world around them, but these representations are either incomplete or flawed. How can we get robots to understand these representations to better assist …
Jupyter Notebook 1
-
plasticity-rl
plasticity-rl PublicDeep RL faces the problem of plasticity loss, which is the loss of ability of neural networks to adapt to new data/conditions. We discover that regularizing singular values is a promising direction…
-
pgasawa/origin
pgasawa/origin PublicOrigin is a browser extension that uses LLMs to redefine the browser experience.
-
If the problem persists, check the GitHub status page or contact support.
