Tight Time Complexities in Parallel Stochastic Optimization with Arbitrary Computation Dynamics

Authors: Alexander Tyurin

ICLR 2025 | Venue PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Leveraging this model and new proof techniques, we discover tight lower bounds that apply to virtually all synchronous and asynchronous methods... The proofs of the lower bounds are based on new proof techniques and constructions...
Researcher Affiliation Academia Alexander Tyurin AIRI, Moscow, Russia Skoltech, Moscow, Russia EMAIL
Pseudocode Yes Method 3 Rennala SGD (Page 6), Method 4 Malenia SGD (Page 7), Algorithm 5 Resisting allocation of the functions {hj} (Page 27), Algorithm 6 Markov process in the jth block (Page 29).
Open Source Code No The paper does not contain any explicit statements about code availability, links to repositories, or mentions of code in supplementary materials.
Open Datasets No The paper is theoretical and does not describe or use any specific datasets for empirical evaluation. It refers to abstract problem setups and theoretical scenarios.
Dataset Splits No The paper is theoretical and does not conduct experiments using datasets, therefore no dataset split information is provided.
Hardware Specification No The paper is theoretical and does not describe any experimental setup or specific hardware used for running experiments. It refers to hardware components like CPUs/GPUs in a general, abstract context related to its theoretical model.
Software Dependencies No The paper is theoretical and does not describe any software implementations or dependencies with specific version numbers.
Experiment Setup No The paper is theoretical and does not describe any experimental setup, including hyperparameters or system-level training settings. It focuses on defining theoretical models and proving complexity bounds.