Personal-Programs using purely python/numpy for better understanding (No external libraries such as Tensorflow, Theano etc. is used in these programs). Here, my research paper is also attached Hybrid_DDPG_paper.pdf
In this repository, i am giving following implementations.
Supervised Learning:
-
Implememtation of paper "Weight Uncertainty in Neural Networks" https://arxiv.org/abs/1505.05424 Bayes_Linear_regression_classification: Here i implemented for Sine curve and MNIST dataset.
-
Implementation of Linear Regression task on XOR truth table and Dummy data (File name: Regression_sigmoid_linear.ipnb), and Classification tasks on MNIST (File name: Vanilla_Neural_Network_Classification_Task_MNIST.ipynb) using Python/Numpy
Reinforcement Learning Algorithms:
- Deep Deterministic Policy Gradient Algorithm (File Name: DDPG.ipynb) Implementation of paper : "Continuous control with deep reinforcement learning" https://arxiv.org/abs/1509.02971