Paper-Implementation

This repository contains implementation of some interesting algorithms mainly from the domain of machine learning, deep learning and reinforcement learning.

Implemented research papers

1.Human-level Control Through Deep Reinforcement Learning (Mnih et al. 2015)

Source: research paper

Description: Using DQN to train an RL agent to play two ATARI games: Pong and Breakout.

2.Action Recognition using Visual Attention (Sharma et al. 2016)

Source: research paper

Description: Soft attention based model for the task of action recognition in videos. I have additionally implemented parts of this research paper in this repository.

3.Mastering the Game of Go without Human Knowledge ( Silver et al. 2017)

... in progress ...