Rethinking Machine Unlearning for Large Language Models

Disclaimer: This is part of my notes on AI research papers. I do this to learn and communicate what I understand. Feel free to comment if you have any suggestion, that would be very much appreciated.

May 31, 313112 · 1 min · Àlex Pujol Vidal

Notes on KAN: Kolmogorov-Arnold Networks

Disclaimer: This is part of my notes on AI research papers. I do this to learn and communicate what I understand. Feel free to comment if you have any suggestion, that would be very much appreciated. This month a new topic hit AI community. A new proposal to find an alternative to Multi-Layer Perceptrons (MLPs) was unleashed, leading to a heavy discussion in social media. Although authors claim that Kolmogorov-Arnold Networks (KANs), the experiments that provide the empirical evidence fall short in complexity....

May 9, 9095 · 7 min · Àlex Pujol Vidal

Lecture Notes on Nonequilibrium Statistical Mechanics

Disclaimer: This is part of my notes on AI research papers. I do this to learn and communicate what I understand. Feel free to comment if you have any suggestion, that would be very much appreciated. Here you can find my notes from the lecture on Nonequilibrium Statistical Mechanics by Chris Jarzynski from University of Maryland. His lecture is available on YouTube: Nonequilibrium Statistical Mechanics - Part 1 Nonequilibrium Statistical Mechanics - Part 2...

April 29, 29299 · 1 min · Àlex Pujol Vidal

Notes on Slide in Defense of Smart Algorithms Over Hardware Acceleration for Large Scale Deep Learning Systems

Disclaimer: This is part of my notes on AI research papers. I do this to learn and communicate what I understand. Feel free to comment if you have any suggestion, that would be very much appreciated. The following post is a comment on the paper SLIDE: In Defense of Smart Algorithms over Hardware Acceleration for Large-Scale Deep Learning Systems by Beidi Chen, Tharun Medini, James Farwell, Sameh Gobriel, Charlie Tai and Anshumali Shrivastava, from Rice University and Intel Corporation....

April 21, 21216 · 7 min · Àlex Pujol Vidal

Notes on The Era of 1-Bit LLMs: All Large Language Models Are in 1.58 Bits

Disclaimer: This is part of my notes on AI research papers. I do this to learn and communicate what I understand. Feel free to comment if you have any suggestion, that would be very much appreciated. The following post is a comment on the paper The Era of 1-Bit LLMs: All Large Language Models Are in 1.58 Bits by Shuming Ma, Hongyu Wang, Lingxiao Ma, Lei Wang, Wenhui Wang, Shaohan Huang, Li Dong, Ruiping Wang, Jilong Xue, and Furu Wei....

April 13, 13133 · 3 min · Àlex Pujol Vidal