Tackling colourization of black and white images with CNNs.
Use Deep Learning to reduce noise in computer generated imagery.
Develop a dynamic pipeline for efficient LLM development.
Evolving teamwork between virtual creatures.
Recommend fashion fits with Neutral Collaborative Filtering.
Understanding hashing in the context of Implicit Neural Representations.
Understand performance spread of protein function prediction models.
Use Quantum ML to preserve quality of sparse signals.
Use sequence modelling and DRL to create a real time trading system.
Build a spiking neural network with resistance RAM.
Our Projects in 2022-2023
Providing an ML solution for the turbine audibility problem to Aercoustics Ltd.
Music Synthesis Using Diffusion Models.
Attempting to beat the Numerai weekly leaderboards.
Reimagining Karl Sims 1994 'Evolving Virtual Creatures' paper.
Generating Osu beatmaps using Seq2Seq models.
Continuation of 2021’s WallStreetBots with a twist -- crypto price prediction.
Advancing the RealTime project from 2021.
Reconstructing the 12-lead ECG information from only 2-3 lead signals.
Bringing information retrieval into the mechanics of creation.
An ML pipeline for the construction industry.
Applying machine learning to picture selection.
Our Projects in 2021-2022
Our Projects in 2020-2021
Our projects in 2019-2020
Our projects in 2018-2019
Price prediction and portfolio/market analysis using NLP and reinforcement learning.
Real estate price prediction on a real-time dataset.
An application of deep convolutional neural network model to specify cardiovascular diseases via analyzing ECG
Using ML to identify and combat employee attrition factors
Recognizing Handwritten Text in Physical Documents
A web application that uses machine learning to detect and rate smiles
A Guide to Brainstorming and Leading a Project.
A Research Project on Benchmarking the Latest Generative Model Based OoOD Methods.
Reproducing "Siamese Neural Networks for One-shot Image Recognition" and benchmarking implementation performance and results.
Developing hand gesture detection for controlling web browsing.
Transfer learning in the domain of musical styles.
Real estate price prediction and transfer learning with CNNs.
Predicting stock prices.
An applied ML NLP project designed to win Cards Against Humanity, a popular humor-based party game. **[Full]**
The UofT Machine Intelligence Student Team Runner (MISTR) is a Discord bot serving the UTMIST Discord server.
Wrapper written around SchedJulia to schedule interviews.