State-of-the-art Research of Deep Reinforcement-learning
- An interset on deep reinforcement-learning research
Hello I am Nitsan Soffair, a Deep RL researcher at BGU.
In my State-of-the-art Research of Deep Reinforcement-learning course you will get the newest state-of-the-art Deep reinforcement-learning research knowledge.
You will do the following
Get state-of-the-art research knowledge regarding
Validate your knowledge by answering short quizzes of each lecture.
Be able to complete the course by ~2 hours.
Advanced exploration methods
Chatbot based Deep RL
Advanced RL metrics
Navigating robot get human language instructions
Merging on-policy and off-policy gradient estimation
More advanced topics
Emergent Tool Use from Multi-Agent Interaction
Emergent Complexity via Multi-Agent Competition
Competitive Self-Play Better Exploration with Parameter Noise
Proximal Policy Optimization
Evolution Strategies as a Scalable Alternative to Reinforcement Learning
Recurrent Experience Reply in distributed Reinforcement-learning
Maximum a Posteriori Policy Optimization
NeuPL: Neural Population Learning
Learning more skills through optimistic exploration
When should agents explore?
Google brain research
QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
FollowNet: Robot Navigation by Following Natural Language Directions with Deep Reinforcement Learning
Interpolated Policy Gradient: Merging On-Policy and Off-Policy Gradient Estimation for Deep Reinforcement Learning
Scalable Deep Reinforcement Learning Algorithms for Mean Field
Value-Based Deep Reinforcement Learning Requires Explicit Regularisation
Air Learning: a deep reinforcement learning gym for autonomous aerial robot visual navigation
Deep Reinforcement Learning at the Edge of the Statistical Precipice
Exploration in Reinforcement Learning with Deep Covering Options
Deep Reinforcement-learning for Dialogue Generation
Who this course is for:
- Anyone who interset on deep reinforcement-learning research
Currently Deep RL researcher at BGU with Masters of CS at BGU.
My thesis topic is Single agent to multi agent (SA2MA) Deep MARL algorithm beats famoues WQMIX created by Shimon whiteson, Head of Waymo reasearch.
My main interest is AI, while I am very enthusiastic about the new research at NLP decided to start teaching as best way for learning.
I have 2 years experience of teaching-assistant at BGU, particularly in Reinforcement learning course.