Khalid Saifullah

Hi.

I'm Khalid - an incoming CS Ph.D. student at the University of Maryland, College Park. I take great interest in Deep Learning research. Specifically, I'm interested in self-supervised representation learning and deep generative modeling. I care about making models broadly applicable by ensuring robustness, sample efficiency and interpretability. Other than that, I'm also into visualizations, photography, aaand cinema!

Latest Posts

Read all posts

Intuitions for the Simpson's Paradox

26.05.2021

Delta Analytics Global Teaching Fellow 2021

09.05.2021

Neural Networks From Linear Algebraic Perspective

07.03.2021

Talks & Workshops

DALL·E mini: Zero-Shot Text-to-Image Generation: A talk I've given as part of my Delta Analytics graduation lecture over YouTube live.
Let’s easily Gittu! Git Github Workshop by UIU APP FORUM: A workshop I've conducted as an instructor for the CS freshers.
OOP Workshop on Java: A 5-day long workshop that I've conducted as an instructor in the trimester break, taught the basics of OOP to the students.

Awards & Community Contributions

Secured 1st position at Huggingface JAX/FLAX Community Event with the group project DALL·E mini.
Received Global Teaching Fellowship by Delta Analytics for 2021 session.
University Merit Scholarship for academic excellence in all trimesters, UIU.
One of the contributors in the Bengali translation of Yann LeCun and Alfredo Canziani's Deep Learning course at NYU.

Projects

DALL·E mini: A JAX/FLAX implementation of OpenAI’s DALL·E but ~30 times smaller and trained on a much smaller dataset to meet resource constraints. (source code)
GPT-2 Bengali: The very first implementation of GPT-2 for Bengali. Trained from scratch with the Bengali portion of multilingual-c4 dataset.(source code)
Bengali Lyricist: A deep generative model that generates Bengali song lyrics. The model is finetuned on the song lyrics dataset from kaggle. (source code)
BERTify: A python module that helps with extracting BERT embeddings for large Benglai/English dataset. Especially optimized for users with limited computational access (e.g. free colab/kaggle GPUs) to tackle CUDA OOMs. (source code)
DietNeRF: First coding implementation of Putting NeRF on a Diet paper that requires much fewer image samples than original NeRF model to perform 3D reconstruction. (source code)
Misogynistic Meme Classifier: As part of AllenNLP Hacks 2021, trained a multimodal model for misogynous meme identification using pretrained CLIP as feature extractor. (source code)
sahajBERT: The very first experiment of decentralized collaborative training with an Albert-like model for Bengali. (source code)
Path Finder: A web app that helps to visualize the typical graph searching algorithm such as Dijkstra, A*, etc. (source code)
8-Puzzle Solver: The 8-puzzle game with the ability to get it solved by the (A* heuristics based) A.I, made as part of the AI course project. (source code)
BOOKy: A website for book lovers to lend and borrow books independently. Made with Django as part of System Analysis and Design course project. (source code)
One Step: A website that helps user to create a social cause and also gets people engage in them as volunteers or donators. Made as part of Database Management System course project. (source code)

See more of my projects on Github

MOOCs & Certifications

Deep Learning Summer School by Neuromatch Academy (certificate) (projects)
Computational Neuroscience Summer School by Neuromatch Academy (certificate)
Harvard CS50's AI course (certificate) (projects)
Codeacademy's Data Science career path (certificate) (projects)

Connect Me

You can reach me at - ksaifullah172043@bscse.uiu.ac.bd