Marco Mondelli
29 papers · 2018–2025 · 6 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+8 more ↓ Show less ↑
π Conference Polyglot (6) π Interdisciplinary Bridge πΊοΈ Taxonomy Completionist (13) π§ Keyword Pioneer π Academic Marathon (7)
π
Academic Marathon
(7)
π
Cross-Pollinator
(11)
π
Keyword Champion
(2)
β‘
Prolific Year
(8)
ποΈ
Keyword Collector
(94)
π₯
Unstoppable
(8)
β
The Questioner
(3)
π
Century Club
(29)
Conferences
ICML (11)
NIPS (9)
COLT (4)
AISTATS (2)
ICLR (2)
JMLR (1)
Top co-authors
Keywords
gradient descent
(7)
approximate message passing
(6)
deep neural network
(4)
spectral method
(3)
generalized linear model
(3)
stochastic gradient descent
(3)
neural tangent kernel
(3)
asymptotic analysis
(2)
neural network optimization
(2)
eigenvalue bound
(2)
feature learning
(2)
neural collapse
(2)
dropout stability
(2)
representation learning
(2)
signal estimation
(2)
rotational invariance
(2)
high-dimensional statistics
(1)
neural network theory
(1)
binary classification
(1)
loss landscape
(1)
Papers
High-dimensional Analysis of Knowledge Distillation: Weak-to-Strong Generalization and Scaling Laws
ICLR 2025
Spurious Correlations in High Dimensional Regression: The Roles of Regularization, Simplicity Bias and Over-Parameterization
ICML 2025
Test-Time Training Provably Improves Transformers as In-context Learners
ICML 2025
Spectral Estimators for Multi-Index Models: Precise Asymptotics and Optimal Weak Recovery
COLT 2025
Wide Neural Networks Trained with Weight Decay Provably Exhibit Neural Collapse
ICLR 2025
Neural Collapse Beyond the Unconstrained Features Model: Landscape, Dynamics, and Generalization in the Mean-Field Regime
ICML 2025
Contraction of Markovian Operators in Orlicz Spaces and Error Bounds for Markov Chain Monte Carlo (Extended Abstract)
COLT 2024
Spectral Estimators for Structured Generalized Linear Models via Approximate Message Passing (Extended Abstract)
COLT 2024
Neural collapse vs. low-rank bias: Is deep neural collapse really optimal?
NIPS 2024
Matrix Denoising with Doubly Heteroscedastic Noise: Fundamental Limits and Optimal Spectral Methods
NIPS 2024
Average gradient outer product as a mechanism for deep neural collapse
NIPS 2024
How Spurious Features are Memorized: Precise Analysis for Random and NTK Features
ICML 2024
Compression of Structured Data with Autoencoders: Provable Benefit of Nonlinearities and Depth
ICML 2024
Towards Understanding the Word Sensitivity of Attention Layers: A Study via Random Features
ICML 2024
Deep Neural Collapse Is Provably Optimal for the Deep Unconstrained Features Model
NIPS 2023
Beyond the Universal Law of Robustness: Sharper Laws for Random Features and Neural Tangent Kernels
ICML 2023
Fundamental Limits of Two-layer Autoencoders, and Achieving Them with Gradient Methods
ICML 2023
Mean-field Analysis of Piecewise Linear Solutions for Wide ReLU Networks
JMLR 2022
Memorization and Optimization in Deep Neural Networks with Minimum Over-parameterization
NIPS 2022
The price of ignorance: how much does it cost to forget noise structure in low-rank matrix estimation?
NIPS 2022
Estimation in Rotationally Invariant Generalized Linear Models via Approximate Message Passing
ICML 2022
When Are Solutions Connected in Deep Networks?
NIPS 2021
PCA Initialization for Approximate Message Passing in Rotationally Invariant Models
NIPS 2021
Tight Bounds on the Smallest Eigenvalue of the Neural Tangent Kernel for Deep ReLU Networks
ICML 2021
Approximate Message Passing with Spectral Initialization for Generalized Linear Models
AISTATS 2021
Global Convergence of Deep Networks with One Wide Layer Followed by Pyramidal Topology
NIPS 2020
Landscape Connectivity and Dropout Stability of SGD Solutions for Over-parameterized Neural Networks
ICML 2020
On the Connection Between Learning Two-Layer Neural Networks and Tensor Decomposition
AISTATS 2019
Fundamental Limits of Weak Recovery with Applications to Phase Retrieval
COLT 2018