Academic Log | August/September 2022
A collection of academic papers/blogs/talks/projects that I read/watched/explored during the month. I also include any small (or large) personal projects that I did and any such related ML/non-ML work.
Personal Projects
- VAE-Implementation - A simple implementation of Autoencoder and Variational Autoencoder - [Github]
- MinHash-Implemenation - A simple MinHash implementation based on the explanation in the Mining of Massive Datasets course by Stanford - [Github]
- Paper re-implementation - Sentence VAE paper, “Generating Sentences from a Continuous Space” by Bowman et al., 2016 - [Github]
- Protohackers - Started the Protohackers set of challenges to create servers for network protocols [Website] [Solutions]
Annotated Papers
- Large-Scale High-Precision Topic Modeling on Twitter
- BadNets - Identifying Vulnerabilities in the Machine Learning Model Supply Chain
- Detecting AI Trojans Using Meta Neural Analysis
- Trojaning Attack on Neural Networks
Papers I read
- Rethinking personalized ranking at Pinterest
- On the Factory Floor: ML Engineering for Industrial-Scale Ads Recommendation Models
- Prioritized Training on Points that are Learnable, Worth Learning, and Not Yet Learnt
- Large-Scale High-Precision Topic Modeling on Twitter
- Detecting AI Trojans Using Meta Neural Analysis
- Trojaning Attack on Neural Networks
- BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
- Neural Cleanse: Identifying and Mitigating Backdoor Attacks in Neural Networks
Blogs I read
- Transformer Precision Loss + Neel Nanda’ Twitter thread on the same
- Deep Q-Networks Explained - LessWrong
- Learning How to Learn - LessWrong
- Making Friends - Neel Nanda’s Blog
- Inside Views - Neel Nanda’s Blog
- Learning - Neel Nanda’s Blog
- MinHash Tutorial
- Alignment Papers Roundup (week 1) - LessWrong
- Intro to JIT - Kipply’s Blog
- Chinchilla’s Wild Implications - LessWrong
- TorchScript - Tracing vs Scripting
- Mastering TorchScript
- Loading a TorchScript Model in C++
- Neural Trojans
Podcasts I listened to
- Engineering an ML-Powered Developer First Search Engine with Richard Socher
- Spotify’s Gustav Söderström on machine learning to personalize user experiences
- George Netscher of SafelyYou on the role of AI for fall detection
- Andrew Song of Whisper AI on solving hearing loss with AI
- How Wayve is teaching cars to drive
- David Rolnick on how machine learning can help tackle climate change
- Mike Fisher of Etsy talks AI and E-commerce
- It’s All About the Data - NerdOut@Spotify
Talks/Videos I watched
- How We’re Reverse Engineering the Human Brain in the Lab | Sergiu P. Pasca | TED
- Creating Personalized Listening Experiences with Spotify
- But what is the Fourier Transform? A visual introduction
- The weirdest paradox in statistics (and machine learning)
- Continual Learning - Full Stack Deep Learning