Zur Hauptnavigation wechseln Zur Suche wechseln Zum Hauptinhalt wechseln

Modern Hopfield Networks

Aktivität: Vortrag oder PräsentationAnderer Vortrag oder PräsentationScience-to-science

Beschreibung

We propose a new paradigm for deep learning by equipping each layer of a deep learning architecture with modern Hopfield networks. The new paradigm is a new powerful concept comprising functionalities like pooling, memory, and attention for each layer. Associative memories date back to the 1960/70s and became popular through Hopfield Networks in 1982. Recently, we saw a renaissance of Hopfield Networks, the modern Hopfield Networks, with a tremendously increased storage capacity and an extremely fast convergence. We generalize modern Hopfield Networks with exponential storage capacity to continuous patterns. Their update rule ensures global convergence to local energy minima and they converge in one update step with exponentially low error. Surprisingly, the transformer attention mechanism is equal to the update rule of our new modern Hopfield Network with continuous states. The new modern Hopfield network can be integrated into deep learning architectures as layers to allow the storage of and access to raw input data, intermediate results, or learned prototypes. These Hopfield layers enable new ways of deep learning, beyond fully-connected, convolutional, or recurrent networks, and provide pooling, memory, association, and attention mechanisms. We demonstrate the broad applicability of the Hopfield layers across various domains. Hopfield layers improved state-of-the-art on three out of four considered multiple instance learning problems as well as on immune repertoire classification with several hundreds of thousands of instances. On the UCI benchmark collections of small classification tasks, where deep learning methods typically struggle, Hopfield layers yielded a new state-of-the-art when compared to different machine learning methods. Finally, Hopfield layers achieved state-of-the-art on two drug design datasets.
Zeitraum22 Sep. 2021
EreignistitelSummer School on "Machine Learning Frontiers in Precision Medicine"
VeranstaltungstypSonstiges
OrtÖsterreichAuf Karte anzeigen

Wissenschaftszweige

  • 101031 Approximationstheorie
  • 102 Informatik
  • 305901 Computerunterstützte Diagnose und Therapie
  • 102033 Data Mining
  • 102032 Computational Intelligence
  • 101029 Mathematische Statistik
  • 102013 Human-Computer Interaction
  • 305905 Medizinische Informatik
  • 101028 Mathematische Modellierung
  • 101027 Dynamische Systeme
  • 101004 Biomathematik
  • 101026 Zeitreihenanalyse
  • 202017 Embedded Systems
  • 101024 Wahrscheinlichkeitstheorie
  • 305907 Medizinische Statistik
  • 102019 Machine Learning
  • 202037 Signalverarbeitung
  • 102018 Künstliche Neuronale Netze
  • 103029 Statistische Physik
  • 202036 Sensorik
  • 202035 Robotik
  • 106005 Bioinformatik
  • 106007 Biostatistik
  • 101019 Stochastik
  • 101018 Statistik
  • 101017 Spieltheorie
  • 101016 Optimierung
  • 102001 Artificial Intelligence
  • 101015 Operations Research
  • 102004 Bioinformatik
  • 101014 Numerische Mathematik
  • 102003 Bildverarbeitung

JKU-Schwerpunkte

  • Digital Transformation