Zafarali Ahmed

All projects

Show all projects from...
data-science | research | software | talks | random
(clear)

Understanding the impact of entropy on policy optimization / [research] / 2017-11-09 00:00:00 +0000
Some work at Google Brain.
An Introduction to Attention Mechanisms / [research] / 2017-11-09 00:00:00 +0000
Slides from my talk at the Montreal Deep Learning Meetup
How to Visualize Your Recurrent Neural Network with Attention in Keras / [research] / 2017-06-29 00:00:00 +0000
A technical discussion and tutorial. Completed during my internship at Datalogue.
Genetic Diversity In Circulating Tumor Cell Clusters / [research] / 2017-03-03 00:00:00 +0000
Genetic diversity plays a central role in tumor progression, metastasis, and resistance to treatment. Using recent progress in numerical models, we simulate macroscopic tumors to investigate the interplay between global growth dynamics, microscopic composition, and circulating tumor cell cluster diversity.
The Rise of Conversational AI / [talk] / 2016-11-01 00:00:00 +0000
I gave a talk at the first Montreal Bot Meetup held at Microsoft HQ. We talked about why bots are great and my journey making one.
Disease Outbreak on College Networks / [research] / 2016-07-01 00:00:00 +0000
Explored the relationship of segregation on a stochastic campus network to the spread and control of a disease.
Minerva Bot / [software] / 2016-05-01 00:00:00 +0000
Created a pseudointelligent chat bot on Facebook Messenger to aid students of McGill find information about courses, programs, buildings and printers faster.
Predicting with Data Workshop / [talks] / 2016-05-01 00:00:00 +0000
Held hands-on workshop to teach the basic concepts behind machine learning and how to implement them using Scikit-Learn.
Comparison of Game Playing Algorithms and Heuristics in Hus / [research] / 2016-04-01 00:00:00 +0000
The exploration of the creation of an agent that plays the game of Hus. The best agent is based on minimax with alpha-beta pruning and a naive heuristic.
Analysis of "Application of quantitative models from population biology and evolutionary game theory to tumor therapeutic strategies" / [research] / 2016-04-01 00:00:00 +0000
Investigation into evolving drug therapies via evolutionary game theory (MATH 437 paper.)
Investigating Evolving Tumours Against Drug Therapies / [talks] / 2016-04-01 00:00:00 +0000
Investigation into evolving drug therapies via evolutionary game theory (MATH 437 talk.)
Towards electroencephalography-based prosthetic control / [research] / 2015-12-01 00:00:00 +0000
Framework and prototype for a Siri-like system that can automatically detect tasks like grasping and liftingthrough the EEG brain signals potentially leading to building better neuroprosthetics.
AngelHack Interview / [talks] / 2015-09-17 00:00:00 +0000
Watch me talk about QuantiScience and the data-driven world.
Computational Oncology / [research] / 2015-07-10 00:00:00 +0000
I simulate cancer growth and their genomes in silico and used it to determine tumor heterogeneity. Over 7000 lines of pure python.
Stressless / [software] / 2015-06-01 00:00:00 +0000
Quantifies stress levels in the workplace using heart rate measurements from commercial wearables like the FitBit. Wrote the algorithm to calculate the Heart Rate Variability.
Flaneur / [software] / 2015-06-01 00:00:00 +0000
A companion travel app that recommends day itineraries based on mood.
Explore Intelligence / [data-science] / 2015-06-01 00:00:00 +0000
Repository being used to learn Machine Learning, Statistics and other skills related to the data driven world.
Simulating Detachment of Tumor Cell Clusters / [research] / 2015-04-10 00:00:00 +0000
TwitFem Attitude Analysis / [data-science] / 2015-04-01 00:00:00 +0000
Used 1M tweets to determine the most characteristic words used by feminisits and anti-feminists with the log-likelihood keyness method.
Travelling Canadians / [data-science] / 2015-03-01 00:00:00 +0000
Bored during reading week, I used data from StatsCan to discover at what age Canadians travel most often throughout their lifetimes.
AngularJS Tutorial / [software] / 2014-06-16 00:00:00 +0000
GitbHub respositiory I used to learn AngularJS to land my first software job. Now has 600+ stars.