Sapana Chaudhary

meta reinforcement learning (2) large language model (2) online convex optimization (2) reinforcement learning from human feedback (1) demonstration learning (1) projected gradient descent (1) risk aversion (1) safe reinforcement learning (1) regret bound (1) linear constraint (1) safety constraint (1) supervised fine-tuning (1) sparse reward (1) educational ai (1) ai alignment (1) conditional value at risk (1) toxicity mitigation (1) offline datum (1) sparse feedback (1) pedagogical alignment (1)