Neural network-based character recognition using MATLAB. The algorithm does not rely on external ML modules, and is rigorously defined from scratch. A report is included which explains the theory, algorithm performance comparisons, and hyperparameter optimization.

Implement of SVM using Stochastic gradient descent - Stony Brook CSE512 Machine learning

Code in MATLAB for 1st order optimization algorithms implemented for elastic net regularized convex objective functions.

Images have to be entered in the database using Image Data Collection Program

Classify 6 types of human activity from time series sensor data

A Single Layer Neural Network to recognize digits making use of unconstrained, non-linear optimization

Single layer perceptron to process images of alphabet letters

Cat vs. car detection with multi-layer NN

Repository contains my MATLAB files for the hand-coded MNIST (w/ SGD optimizer) classification model trained for the EEL5813 - Neural Networks: Algorithms and Applications course, PROJECT02

Stochastic Gradient Descent implementation for SoftSVM

Implemented Kernel SVM using Quadratic Programming and Stochastic Gradient Descent

Train a SVM and for detecting human upper bodies in TV series The Big Bang Theory

MATLAB implementations of a variety of machine learning/signal processing algorithms.

Subsampled Riemannian trust-region (RTR) algorithms

Simple MATLAB toolbox for deep learning network: Version 1.0.3

AAAI & CVPR 2016: Preconditioned Stochastic Gradient Langevin Dynamics (pSGLD)

OnLine Low-rank Subspace tracking by TEnsor CP Decomposition in Matlab: Version 1.0.1

Riemannian stochastic optimization algorithms: Version 1.0.3

Matlab implementation of the Adam stochastic gradient descent optimisation algorithm