Interactive Visual Machine Learning Demos.

[EMNLP'20][Findings] Official Repository for the paper "Why and when should you pool? Analyzing Pooling in Recurrent Architectures."

Deep Neural Networks for music genre classification as a proxy for multiple analytical studies

Pipeline for solving the Vanishing Gradient problem in Deep Learning. Implements methods such as ReLU, BatchNorm, Skip Connection, and gradient optimization to improve deep model training.

Multilayer Perceptron GAN, and two Convolutional Neural Network GANs for MNIST and CIFAR.

This repository helps in understanding vanishing gradient problem with visualization

Adaptive-saturated RNN: Remember more with less instability

Code repository for my CSU master's research on dead ReLU's

This pdf shows some methods to fine tune the hyperparameters of a Neural Network to increase it's performance. It also puts light on some of the common problems related to Neural Networks along with their solutions.

Python codes for the experiment about LSTM Inefficiency in Long-Term Dependencies Regression Problems

This repository is based on the discussion about the Vanishing Gradient Problem. It explains some of the causes of this issue and provides solutions to help mitigate it.

Code for NeurIPS 2024 paper "Only Strict Saddles in the Energy Landscape of Predictive Coding Networks?"

The vanishing gradient problem is a well-known issue in training recurrent neural networks (RNNs). It occurs when gradients (derivatives of the loss with respect to the network's parameters) become too small as they are backpropagated through the network during training.

Machine Learning Practical - Coursework 2 Report: Analysing problems with the VGG deep neural network architectures (with 8 and 38 hidden layers) on the CIFAR100 dataset by monitoring gradient flow during training. And exploring solutions using batch normalization and residual connections.

I'll try to explain through the outcomes of Vanishing Gradient Problem

Everything about Artificial Neural Network from Basic to Adavnced

Vanishing Gradient and Activation Functions

An interactive 3D "Glass Box" sandbox built with pure NumPy and WebGL to visualize neural network math, loss trajectories, and decision boundaries in real-time.

Machine Learning Practical - Coursework 2: Analysing problems with the VGG deep neural network architectures (with 8 and 38 hidden layers) on the CIFAR100 dataset by monitoring gradient flow during training. And exploring solutions using batch normalization and residual connections.