AutoenCODE is a Deep Learning infrastructure that allows to encode source code fragments into vector representations, which can be used to learn similarities.

Extract features and detect anomalies in industrial machinery vibration data using a biLSTM autoencoder

Auto-encoder Based Data Clustering Toolkit

Code for paper "Autoencoder Inspired Unsupervised Feature Selection"

Source code for 3D volumetric denoising auto-encoder (ECCV-16)

Cost function and cost gradient function for a convolutional autoencoder.

UB Computer Vision

The main goal of this work is to build and train multilayer NNs, train autoencoders to reduce the number of features for the classifiers and build and train deep networks (CNN and LSTM) for predicting or detecting the seizures.

Real-world application sized Neural Network. Implemented back-propagation algorithm with momentum, auto-encoder network, dropout during learning, least mean squares algorithm.

My solutions to UFLDL Tutorial (http://deeplearning.stanford.edu/wiki/index.php/UFLDL_Tutorial)

Some of the work I completed for CS/Math 350: Mathematical Modeling

Code for the analysis conducted in the paper "On the Importance of Hidden Bias and Hidden Entropy in Representational Efficiency of the Gaussian-Bipolar Restricted Boltzmann Machines"