Maksim Velikanov
7 papers · 2021–2025 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+1 more ↓ Show less ↑
๐ Interdisciplinary Bridge ๐งญ Keyword Pioneer ๐ Conference Polyglot (4) ๐ Cross-Pollinator (15) ๐บ๏ธ Taxonomy Completionist (14)
๐ฅ
Unstoppable
(5)
Conferences
ICLR (3)
AISTATS (2)
JMLR (1)
NIPS (1)
Top co-authors
Keywords
gradient descent
(2)
neural network
(2)
stochastic gradient descent
(1)
conformal prediction
(1)
ensemble learning
(1)
uncertainty quantification
(1)
neural network optimization
(1)
spectral analysis
(1)
conjugate gradient
(1)
weight sharing
(1)
quadratic optimization
(1)
convergence rate
(1)
data heterogeneity
(1)
confidence set
(1)
momentum method
(1)
gradient descent dynamics
(1)
embedded ensemble
(1)
infinite width limit
(1)
dropout ensemble
(1)
heavy ball method
(1)
Papers
SGD with memory: fundamental properties and stochastic acceleration
ICLR 2025
Generalization error of spectral algorithms
ICLR 2024
Tight Convergence Rate Bounds for Optimization Under Power Law Spectral Conditions
JMLR 2024
Efficient Conformal Prediction under Data Heterogeneity
AISTATS 2024
A view of mini-batch SGD via generating functions: conditions of convergence, phase transitions, benefit from negative momenta.
ICLR 2023
Embedded Ensembles: infinite width limit and operating regimes
AISTATS 2022
Explicit loss asymptotics in the gradient descent training of neural networks
NIPS 2021