Implementation of ECO-DQN as reported in "Exploratory Combinatorial Optimization with Reinforcement Learning".

Real-Time Locomotion on Soft Grounds with Dynamic Footprints - 2022 - Frontiers in Virtual Reality

Three-dimensional residual channel attention networks

1-dimensional convolutional neural networks (CNN) for the classification of soil texture based on hyperspectral data

Supporting code for "End-to-end optical backpropagation for training neural networks".

MWF-based EEG artifact removal in MATLAB

Generating Upper-Body Motion for Real-Time Characters Making their Way through Dynamic Environments - 2022 - SCA

Soft Walks: Real-Time, Two-Ways Interaction between a Character and Loose Grounds - 2021 - Eurographics (short)

Publicly available code from my publications

This repository contains scripts from published or submitted manuscripts.

Suite of Jupyter Notebook-based tools for interpreting and analyzing standard and capillary gel data.

Code for paper titled, "BSite-pro: A Novel Approach for Binding Site Prediction in Protein Sequences".

📝 A Gaussian-process-model-based approach for robust, independent, and implementation-agnostic validation of complex multi-variable measurement systems: application to SAR measurement systems

Simulates aperiodic EEG signals and detrends EEG spectra, as well as generates figures used in manuscript.

A collection of custom scripts used for Gaccioli et al. Nat. Microbiol. 2020.

R Code for the paper entitled "Identifying and responding to outlier demand in revenue management". Published in European Journal of Operational Research. 2021.

Reproducible code for manuscript "Venn diagram analysis overestimates the extent of circadian rhythm reprogramming" published in FEBS Journal

The Simulink Model Mutation Testing project provides a framework for generating mutants, and CSV file as an output. It utilizes the MATLAB environment to load the model, apply mutations, and collect outputs.

QuantEase, a layer-wise quantization framework, frames the problem as discrete-structured non-convex optimization. Our work leverages Coordinate Descent techniques, offering high-quality solutions without the need for matrix inversion or decomposition.