Anthill Inside 2019

On infrastructure for AI and ML: from managing training data to data storage, cloud strategy and costs of developing ML models

Reinforcement Learning

Submitted by Suraj Sheth (@shethsh) on Monday, 29 April 2019

Section: Workshops Technical level: Intermediate Session type: Workshop

Abstract

We see quite a lot of excitement around Reinforcement Learning (RL) in industry and academia. We regularly see new industrial applications both inside and outside of Amazon. We also see RL papers dominating many ML conferences such as NeurIPS and ICML. RL is promising for multiple reasons that include a better formulation of the problem which allows for optimization of multi-step decision making process and its ability to learn strategies for unseen, dynamic scenarios. RL has gained traction in recent years with the use of deep neural networks to accomplish astonishing results in applications such as games, robotics, recommendations and operations research problems. We propose this workshop to bring people together to share ideas, increase awareness and build a community around RL.

Outline

The workshop is designed to give an overview of RL and dive deep into some of the state-of-the-art techniques in RL.
Reinforcement Learning:
1) Introduction
2) Definitions and Problem Statement
3) Classical Approaches
4) Evolutionary Algorithms
5) Temporal Differences
6) Q-Learning
7) Exploration and Exploitation
8) Evaluating Reinforcement Learning Algorithms
9) On-Policy Learning
10) Assigning Credit and Blame to Paths
11) Model-Based Methods
12) Reinforcement Learning with Features
13) Deep Reinforcement Learning
14) Atari Games, Self-Driving and other applications

Requirements

Laptop with Python installed/Cloud host

Speaker bio

Suraj is a Senior Machine Learning Scientist working on a variety of Machine Learning problems at Amazon. Prior to this, he has worked at IBM Research(IRL) and Adobe. His areas of work include Natural Language Processing, Deep Recurrent Networks, CNNs for text, Reinforcement Learning, and Generative Models. Suraj has more than 8 years of industry experience in the field of Machine Learning, NLP, Deep Learning and Reinforcement Learning. He has presented papers/posters in AMLC for last three years, delivered talks/tutorials at GHCI, AITC and others. He has three USPTO approved patents. He has conducted Reinforcement Learning workshops at several occasions including Amazon SMML Workshop 2018. He is organizing “Reinforcement Learning” workshop at Amazon Machine Learning Conference 2019.

Comments

  • Abhishek Balaji (@booleanbalaji) Reviewer a month ago

    Hi Suraj,

    Thanks for submitting a proposal. We’re evaluating this and someone from our team will be in touch with you regarding this proposal.

Login with Twitter or Google to leave a comment