Projects🚧

Reinforcement Learning

Policy Gradient Theories

  1. Notes on Policy Gradient Methods Under Review

Policy Gradient Algorithms

  1. DQN - Human-level Control Through Deep Reinforcement Learning (Nature2015) Under Review
  2. DQN - Playing Atari with Deep Reinforcement Learning (NIPS2013) Under Review
  3. DPG - Deterministic Policy Gradient Algorithms (ICML2014) Under Review
  4. DDPG - Continuous Control with Deep Reinforcement Learning (ICLR2016) Under Review
  5. A3C - Asynchronous Methods for Deep Reinforcement Learning
  6. PPO - Proximal Policy Optimization Algorithms
  7. TD3 - Addressing Function Approximation Error in Actor-Critic Methods
  8. SAC - Soft Actor-Critic Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor
  9. HER - Hindsight Experience Replay
  10. HER - Multi-Goal Reinforcement Learning Challenging Robotics Environments and Request for Research

UAV Collision Avoidance

  1. Comparison of Deep Reinforcement Learning Policies to Formal Methods for Moving Obstacle Avoidance
  2. Reinforcement Learning for Autonomous Driving Obstacle Avoidance using LIDAR