Skip to content

RL and DMP algorithms implemented from scratch with plain Numpy.

Notifications You must be signed in to change notification settings

alaradirik/robot-learning

Repository files navigation

Robot Learning

Boğaziçi University, Special Topics in Robot Learning Projects

This repo consists of a set of applications and algorithms implemented from scratch with numpy for robot learning. Included algorithms are tabular q-learning, deep neural network based q-learning, and dynamical movement primitives.

Project Structure

.
├── q_learning.py                   # Tabular Q-learning
├── deep_q_learning.py              # Deep Neural Network based Q-learning
├── policy_gradient.py              # REINFORCE algorithm
├── DMP.ipynb                       # Dynamical Movement Primitives
├── DMP_GMM/                        # Kernelized Dynamical Movement Primitives with GMM-GMR
└── README.md