Benjamin F Grewe

4 papers · 2021–2025 · 2 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (12) 🐣 Hot Topic Early Bird

Venues

ICLR (2) ICML (2)

Top co-authors

Alexander Meulemans (2) Christian Henning (2) Johannes von Oswald (2) Thilo Stadelmann (1) Maria Cervera (1) Bernhard Schölkopf (1) Benjamin Ehret (1) Matteo Saponati (1) Seijin Kobayashi (1) João Sacramento (1)

Research topics

Machine Learning (4) Deep Learning (3) Architectures (3) Core Methods (2) Learning Types · Machine Learning (2) Optimization & Theory (2) Natural Language Processing (1) Application Areas (1) Learning Types · Deep Learning (1) Resources & Methods (1)

Keywords

representation learning (1) continual learning (1) catastrophic forgetting (1) stochastic gradient descent (1) model merging (1) symmetry (1) group theory (1) action prediction (1) recurrent neural network (1) latent space (1) autoregressive training (1) self attention (1) elastic weight consolidation (1) bidirectional training (1) emergent dynamics (1) weight ensembling (1) neural network (1) weight importance (1) group representation (1) directionality (1)

Papers

The underlying structures of self-attention: symmetry, directionality, and emergent dynamics in Transformer training ICML 2025 Homomorphism AutoEncoder -- Learning Group Structured Representations from Observed Transitions ICML 2023 Continual learning in recurrent neural networks ICLR 2021 Neural networks with late-phase weights ICLR 2021