Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/dane-meister/machine-learning-algos

Repository of several ML algorithms implemented for different environments and robot control vectors.
https://github.com/dane-meister/machine-learning-algos

colab-notebook gymnasium jupyter-notebook machine-learning-algorithms numpy opencv pybullet python pytorch

Last synced: 7 days ago
JSON representation

Repository of several ML algorithms implemented for different environments and robot control vectors.

Awesome Lists containing this project

README 

        
# Machine Learning Algorithms for Robotics

Welcome to my repository where I showcase a collection of machine learning algorithms, primarily focused on applications in robotics. This repository is a compilation of assignments and projects I completed during my robotics course at Stony Brook University.



## Overview

In this repository, you'll find various machine learning algorithms designed for different environments and robot control vectors. Each algorithm is thoughtfully implemented and tested, ensuring relevance and efficacy in practical robotics applications.



### Features

- Implementations of diverse machine learning algorithms.

- Application of these algorithms in robotics.

- Use of Python and libraries like PyTorch for robust development.



### Repository Structure

Each algorithm in this repository is organized into a separate Jupyter notebook. These notebooks were created using Google Colab, ensuring ease of access and the ability to run these notebooks with minimal setup.



## Algorithms Included



- **[Nearest Neighbor]**: A simple yet effective algorithm that classifies data points based on the proximity to data points seen during training. Commonly used for its ease of implementation and interpretability in robotic applications.



- **[Linear Regression]**: A foundational algorithm in statistics and machine learning for predicting a continuous outcome variable based on one or more predictor variables. It's especially useful in robotics for understanding and predicting linear relationships.

- **[Multi-Layer Perceptron]**: A class of feedforward artificial neural network that consists of at least three layers of nodes. MLPs are widely used for complex pattern recognition tasks, including speech, image, and motion recognition in robotics.

- **[Monte-Carlo]**: These are computational algorithms that rely on repeated random sampling to obtain numerical results, typically used in robotics for optimization and probabilistic learning in uncertain environments.

- **[Q-Learning]**: A model-free reinforcement learning algorithm used to learn the value of an action in a particular state. It's useful in robotics for making sequences of decisions without a model of the environment.

- **[Deep Q-Network]**: An advanced reinforcement learning algorithm that combines Q-Learning with deep neural networks, allowing the handling of high-dimensional sensory inputs for decision-making in complex robotic tasks.

- **[Reinforce]**: A type of policy gradient method in reinforcement learning, which optimizes the policy directly. It's beneficial in robotics for scenarios where the algorithm needs to learn a complex policy from high-dimensional data.

- **[Actor-Critic]**: These methods utilize both policy and value function to solve reinforcement learning problems. The actor proposes actions given states, and the critic evaluates them, enabling efficient learning in complex robotic control tasks.



### Getting Started

To get started with these notebooks:



**Clone the Repository**



git clone https://github.com/dane-meister/Machine-Learning-Algos.git



**Navigate to the Notebook of Interest**



Each notebook is self-contained with instructions and explanations.



**Run the Notebooks in Google Colab**



You can easily open these notebooks in Google Colab and run them to see the algorithms in action.



###### Contributing

Feel free to fork the repository, make changes, or suggest improvements by creating a pull request. I appreciate any contributions or feedback to enhance the repository.



###### License

This project is open-source and available under the MIT License.



###### Acknowledgments

Stony Brook University CS Department.

Instructors (Michael Ryoo, Jinghuan Shang) and peers who provided insights and feedback.