Berfin Simsek
8 papers · 2020–2025 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+3 more ↓ Show less ↑
🗺️ Taxonomy Completionist (13) 🌉 Interdisciplinary Bridge 🧭 Keyword Pioneer 🌍 Conference Polyglot (4) 🏃 Academic Marathon (5)
🐝
Cross-Pollinator
(4)
📈
Trend Setter
❓
The Questioner
Conferences
ICML (4)
NIPS (2)
AISTATS (1)
ICLR (1)
Top co-authors
Keywords
kernel ridge regression
(2)
learning theory
(1)
knowledge distillation
(1)
loss landscape
(1)
generalization error
(1)
random matrix theory
(1)
gradient flow
(1)
implicit regularization
(1)
weight averaging
(1)
affine subspace
(1)
random feature model
(1)
critical point
(1)
overparameterized neural network
(1)
risk prediction
(1)
kernel alignment
(1)
permutation symmetry
(1)
kernel methods
(1)
statistical learning
(1)
neural network
(1)
Papers
Learning Gaussian Multi-Index Models with Gradient Flow: Time Complexity and Directional Convergence
AISTATS 2025
Loss Landscape of Shallow ReLU-like Neural Networks: Stationary Points, Saddle Escape, and Network Embedding
ICLR 2025
Learning Associative Memories with Gradient Descent
ICML 2024
Expand-and-Cluster: Parameter Recovery of Neural Networks
ICML 2024
Should Under-parameterized Student Networks Copy or Average Teacher Weights?
NIPS 2023
Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances
ICML 2021
Implicit Regularization of Random Feature Models
ICML 2020
Kernel Alignment Risk Estimator: Risk Prediction from Training Data
NIPS 2020