[CW Paper-Club] Hopfield Networks is All You Need

In this week's session, Maruan Ottoni presents the paper "Hopfield Networks is All You Need" published in 2021 by Hubert Ramsauer, Bernhard Schäfl, Johannes Lehner et al. from Johannes Kepler University Linz and University of Oslo.

The paper introduces a modern Hopfield network with continuous states and an equivalent update rule resembling the attention mechanism in transformers. Demonstrating broad applicability, Hopfield layers improve state-of-the-art in multiple instance learning, immune repertoire classification, and small classification tasks. Integrated into deep learning architectures, these layers offer new avenues beyond traditional networks, providing pooling, memory, association, and attention mechanisms.

If you're interested in joining our team at CloudWalk, please check out our job openings on LinkedIn! Don't forget to check out the paper, which is available at the link provided below.

Recording Date: January 19 2024
00:00 Introduction to Hopfield networks
19:00 Deep dive into the paper

CW Paper-Club Sessions is a weekly meeting presented by CloudWalk's R&D team members, where they dive into interesting research papers, providing valuable insights into the latest developments in technology and research.

Full Paper Abstract: