· van Vreeswijk TNS · Talk
A function approximation perspective on neural representations
Harvard University
Computational Neuroscience · Video · 59 min
Cengiz Pehlevan
Harvard University,
1 talk · 13 posters
Recordings
Posters · 13
Mar 27, 2025 · COSYNE 2025 · Poster
Dynamically Learning to Remember in Recurrent Neural Networks
Mar 27, 2025 · COSYNE 2025 · Poster
Equality reasoning in neural networks is modulated by learning richness
Mar 27, 2025 · COSYNE 2025 · Poster
A Model of Place Field Reorganization During Reward Maximization
Mar 27, 2025 · COSYNE 2025 · Poster
Simultaneous detection and mapping in the olfactory bulb
Mar 27, 2025 · COSYNE 2025 · Poster
Mechanistic biases in data-constrained models of neural dynamics
Mar 11, 2023 · COSYNE 2023 · Poster
Emergent compositional reasoning from recurrent neural dynamics
Show 7 more
Mar 11, 2023 · COSYNE 2023 · Poster
Influence of Learning Rules on Representation Dynamics in Wide Neural Networks
Mar 11, 2023 · COSYNE 2023 · Poster
Robust Sequence Recall in Asymmetric Modern Hopfield Networks
Harvard University
Mar 11, 2023 · COSYNE 2023 · Poster
Variance-limited scaling laws for plausible learning rules
Harvard University
Mar 19, 2022 · COSYNE 2022 · Poster
Learning sequences with fast and slow parts
Mar 19, 2022 · COSYNE 2022 · Poster
How many objects can be recognized under all possible views?
Mar 17, 2022 · COSYNE 2022 · Poster
Neural network size balances representational drift and flexibility during Bayesian sampling
Mar 17, 2022 · COSYNE 2022 · Poster