Nelineinaya 2024

4 papers
Nelineinaya 2024 featured research focused on advanced optimization techniques, particularly regarding black-box systems and high-order regularity. A significant portion of the work addressed zeroth-order optimization, exploring its application in overparameterized regimes and under conditions of high-order smoothness. Other prominent themes included the use of comparison oracles for quasi-convex smooth optimization and the development of parameter-free methods through universal adaptive stepsizes. The year's contributions also examined the interpolation regime and the reconciliation of optimization theory with non-convex loss landscapes. Methodological trends included the application of Performance Estimation Problems (PEP) to bilinear-coupled saddle point problems and the asymptotic analysis of Ruppert-Polyak averaging for stochastic order oracles. Research also covered automated performance estimation and optimization under adversarial noise. Prolific contributors to this edition included Alexander Gasnikov, Fedor Stonyakin, and Aleksandr Lobanov, whose work spanned topics from gradient direction normalization to kernel-based zeroth-order optimization for highly smooth functions.

Top Research Directions

The 12 research directions with Nelineinaya 2024 papers, ranked by paper count. Share is the fraction of the edition’s papers in the direction.

#DirectionPapersShare
1Overcoming the Convexity Constraint in Modern Optimization375.0%
2Optimization via Comparison Oracles250.0%
3Zeroth Order Stochastic Optimization250.0%
4Exploiting Higher Order Smoothness in Optimization125.0%
5Finite-Time Bounds for Stochastic Iterative Algorithms125.0%
6Gradient Approximation via Random Function Perturbations125.0%
7How Vanishing Stochastic Noise Accelerates Overparameterized Learning125.0%
8Iterate Averaging for Optimal Stochastic Approximation125.0%
9Overcoming Geometric Mismatch in Parameter Space Optimization125.0%
10Performance Estimation Programming for Optimization Algorithms125.0%
11Semidefinite Relaxation for Non-Convex Engineering Optimization125.0%
12The Mechanics of Benign Overfitting in High Dimensions125.0%

Download all directions as CSV

Institutions

Institutions credited by Nelineinaya 2024 papers: a paper credits every institution named in its affiliations, once. 14 institutions share 4 credited papers.

#InstitutionPapersShare
1Innopolis University 4100.0%
2Moscow Institute of Physics and Technology 375.0%
3Institute for System Programming 250.0%
4Lomonosov Moscow State University 250.0%
5National Research University Higher School of Economics 250.0%
6Skolkovo Institute of Science and Technology 250.0%
700tf0yk25 125.0%
803eeyjx56 125.0%
9AIRI - Artificial Intelligence Research Institute 125.0%
10Bennett University 125.0%
11Oracle (United States) 125.0%
12Russian Academy of Sciences 125.0%
13Steklov Mathematical Institute 125.0%
14V.I. Vernadsky Crimean Federal University 125.0%

Download all institutions as CSV

About this data

Papers. Nelineinaya 2024 lists 4 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 4 of 4 papers (100.0%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-17 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “Nelineinaya 2024”, https://lacuna.tiptreesystems.com/venue/nelineinaya/2024.

Top Authors

AuthorPapers
Aleksandr Lobanov3
Dmitry Kovalev1
Ilya Kuruzov1

Papers

Title
Accelerated Zero-Order SGD under High-Order Smoothness and Overparameterized Regime
Asymptotic Analysis of the Ruppert – Polyak Averaging for Stochastic Order Oracle
Convex-Concave Interpolation and Application of PEP to the Bilinear-Coupled Saddle Point Problem
On Quasi-Convex Smooth Optimization Problems by a Comparison Oracle