Levent Sagun
10 papers · 2018–2025 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
๐ Cross-Pollinator (15) ๐งญ Keyword Pioneer ๐ฃ Hot Topic Early Bird ๐ Conference Polyglot (4) ๐ Academic Marathon (7)
๐
Interdisciplinary Bridge
๐บ๏ธ
Taxonomy Completionist
(16)
๐
Century Club
(10)
โ
The Questioner
Conferences
ICML (4)
NIPS (3)
ICLR (2)
NAACL (1)
Top co-authors
Keywords
loss landscape
(2)
neural network
(2)
statistical physics
(2)
vision transformer
(1)
text classification
(1)
toxicity detection
(1)
neural network optimization
(1)
generalization error
(1)
speech classification
(1)
logistic loss
(1)
convolutional neural network
(1)
heavy-tailed distribution
(1)
stochastic differential equation
(1)
training dynamics
(1)
annotation quality
(1)
saddle point
(1)
translation invariance
(1)
classification problem
(1)
data structure
(1)
interpolation threshold
(1)
Papers
An Effective Theory of Bias Amplification
ICLR 2025
A Differentiable Rank-Based Objective for Better Feature Learning
ICLR 2025
On the Role of Speech Data in Reducing Toxicity Detection Bias
NAACL 2025
Networked Inequality: Preferential Attachment Bias in Graph Neural Network Link Prediction
ICML 2024
ConViT: Improving Vision Transformers with Soft Convolutional Inductive Biases
ICML 2021
On the interplay between data structure and loss function in classification problems
NIPS 2021
Triple descent and the two kinds of overfitting: where & why do they appear?
NIPS 2020
Finding the Needle in the Haystack with Convolutions: on the benefits of architectural bias
NIPS 2019
A Tail-Index Analysis of Stochastic Gradient Noise in Deep Neural Networks
ICML 2019
Comparing Dynamics: Deep Neural Networks versus Glassy Systems
ICML 2018