Manan Tomar

Mirror-Descent-Policy-Optimization. Mirror Descent Policy Optimization

DSR. Deep Successor Representation

MPC. Contains simple MPC implementation with neural network learned dynamics.

video-occupancy-models. Python

agent-centric-representations. Code for "Principled Offline RL in the Presence of Rich Exogenous Information" and "Ignorance is Bliss: Robust Control via Information Gating".

Reinforcement-Learning. Contains implementations of various deep RL algorithms and papers including action conditional video prediction | Python | Tensorflow | Open AI gym

Multi-step-Greedy-Reinforcement-Learning-Algorithms. Multi-step Greedy Reinforcement Learning Algorithms

Monocular_SLAM. Implementation of monocular SLAM using Bundle Adjustment and Key Frame Techniques

test. C++