VeriCache: Making Lossy KV Compression Exact
Long-context inference has made the KV cache one of the main bottlenecks in serving large language models.
Technical writing
Deep dives into machine learning, systems, and technology.
2026
18 min read
2024
build a multiscale CNN with me! step-by-step guide to implement a multiscale CNN!
Implementing a multiscale CNN & understanding the theory / motivation behind building a multiscale CNN.
build a GAN with me! step-by-step guide on understanding a GAN from scratch!
This article goes over what GANs are, how they work, how to implement them from scratch in python & using them on a sample dataset!
build an autoencoder with me! step-by-step guide on understanding autoencoders and training them on MNIST!
An in-depth breakdown of what an autoencoder is, the math / theory behind them, using them for mnist dataset, and the types of…
build GPT with me! implementing GPT from scratch step-by-step
A step by step guide on implementing GPT from scratch in Python.
2023
a picture is worth a 1000 words; breaking down Vision Transformers
breaking down vision transformers with code implementation.
reducing sycophancy in LLMs with synthetic data
How we can use synthetic data to reduce sycophancy in Large Language Models
reconstructing music from brain waves!? meet Brain2Music
This article delves into reconstructing music from brain activity scans with MusicLM!
Saying bye to segmentation models!? Understanding Universal Segmentation
The main idea of this paper is to leverage a diverse collection of segmentation datasets to co-train a single model for segmentation tasks.
why has GPT-4’s accuracy been declining so much!?
Understanding the Perplexing Decline of GPT-4’s Accuracy!
Breaking Down Contrastive Learning with a Hierarchical Twist
Understanding and breaking down the key components of Hierarchical Contrastive Learning.
Detecting COVID-19 with Precision: A Triplet Siamese Neural Net Framework
CovidExpert: A Triplet Siamese Neural Network framework for the Detection of COVID-19
Leveraging Zero-Shot Learning to Detect Pathologies with Unannotated Data
Expert-level detection of pathologies from unannotated chest X-ray images via self-supervised learning
2022
Back-Propagation No More? Breaking Down the Forward-Forward Algorithm
The Forward-Forward Algorithm: Some Preliminary Investigations
Expressing Neural Networks as Decision Trees
Explaining how any Neural Network with an Activation can be Expressed as an Equivalent Decision Tree
Leveraging Federated Learning for Medical Imaging
High-Level Overview of Federated Learning x Medical Imaging
Federated Learning; The Future of Decentralized Machine Learning
A breakdown of a newly researched subset of Machine Learning, Federated Learning.
Breaking Down Natural Language Processing
A step-by-step breakdown of Natural Language Processing.
Off-Wind Energy; Pioneering a New Approach to Renewable Energy
This paper was a collective effort of Pavi Dhiman, Shiza Charania, Bagavan Marakathalingasivam and I
2021
Training a CNN to detect Face Masks
We often take things for granted, there’s always the famous saying
What are Neural Networks?
Imagine this. You’re in a foreign country and all the signs around you are in a language that you don’t understand. The most logical way…
The Future of Food: Cellular Agriculture
23 billion chickens. 1.5 billion cattle. 1 billion pigs. 1 billion sheep. These numbers seem all so insignificant at first, but what if I…
How AR/VR is Going To Change The Future of Alzheimer’s
If I walked up to you and asked for the first word that comes to mind when I say Virtual or Augment reality, most of you would say…
2020
The Drug Discovery Process and The Development of a Covid-19 Vaccine
Drugs are one of the most important if not most important aspect of the pharmaceutical world. However, these drugs don’t simply burst onto…
The Future of Precision Medicine
Over 463 million people have Diabetes, 44 million people suffer from Alzheimer's, 7 million people have Parkinson’s Disease. You may…
No posts found. Try a different search.