Minhajul Hoque

Featured in: Favicon

medium.com

Articles

AI data development: a guide for data science projects

Nov 13, 2024 | snorkel.ai | Matthew Casey |Matt Casey |Minhajul Hoque

What is AI data development? AI data development includes any action taken to convert raw information into a format useful to AI. This definition covers everything from the point of acquiring data to the point at which a data scientist exports a model for deployment.
Insights from LLAMA 3.1 Paper

Jul 26, 2024 | medium.com | Minhajul Hoque

Image from https://www.analyticsvidhya.com/blog/2024/07/meta-llama-3-1/ The release of LLAMA 3.1 marks a significant milestone in AI development. For the first time, an open-source model is approaching the performance levels of leading closed-source models. This shift hints at a future where open-source models are just as effective, providing the flexibility for anyone to modify and adapt them.
How we achieved 89% accuracy on contract question answering

Apr 2, 2024 | snorkel.ai | Minhajul Hoque

Imagine sifting through a 500-page legal document for several hours to answer a simple question. Now, imagine a machine learning system that could perform contract question answering for you in a fraction of the time. That’s what we at Snorkel AI recently created for a top 10 US bank.
How to Become a Machine Learning Engineer

Dec 29, 2023 | medium.com | Minhajul Hoque

As an Aerospace Engineering graduate, my journey into the cosmos of Machine Learning (ML) and Artificial Intelligence (AI) was as unexpected as it was fascinating. Inspired by a podcast featuring AI pioneer Andrew NG, my curiosity pivoted from the universe to the universe of data. This blog is a chronicle of that transition, and a guide for those intrigued by the potential of AI. Armed with curiosity and the internet, my deep dive into AI was intense and exhilarating.
How to Handle The Lack of Labeled Data

Feb 4, 2023 | medium.com | Minhajul Hoque

AI models are excellent at tasks such as image classification, object detection, and sentiment analysis. However, building a strong supervised learning model requires a large amount of labeled data, which can be time-consuming and prone to errors. To handle the issue of limited labeled data, there are several approaches:Weak SupervisionSelf-Supervised LearningTransfer LearningActive LearningWeak SupervisionWeak supervision enables the labeling of unlabeled data using label functions (LFs).