Oracle efficient truncated statistics

Authors: Konstantinos Karatapanis, Vasilis Kontonis, Christos Tzamos

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

Reproducibility Variable Result LLM Response
Research Type Theoretical In this work we design a new learning method with runtime and query complexity polynomial in 1/α. Our result significantly improves over the prior works by focusing on efficiently solving the underlying optimization problem using a general purpose optimization algorithm with minimal assumptions. Our main contribution is a positive answer to the above question by providing a new analysis of the truncated negative log likelihood objective. The parameter θ that will satisfy the properties of the theorem will be the output of the Algorithm 1.
Researcher Affiliation Academia Konstantinos Karatapanis1,2, Vasilis Kontonis3, Christos Tzamos1,4 1 Archimedes, Athena Research Center, Greece 2 National Technical University of Athens 3 University of Texas at Austin 4 University of Athens EMAIL, EMAIL, EMAIL
Pseudocode Yes Algorithm 1 Algorithm for Minimizing L-Constrained Function Algorithm 2 Projected SGD Algorithm Given Truncated Samples and Sublevel Set Algorithm 3 Sample Gradient
Open Source Code No The paper does not contain any explicit statement about open-source code availability or links to code repositories.
Open Datasets No The paper is theoretical and does not conduct empirical studies using specific datasets. Therefore, it does not provide any information about publicly available or open datasets.
Dataset Splits No The paper is theoretical and does not conduct empirical studies. Therefore, it does not provide information about dataset splits.
Hardware Specification No The paper is theoretical and does not report on experiments requiring specific hardware. Therefore, no hardware specifications are provided.
Software Dependencies No The paper is theoretical and does not describe any specific implementation details or list software dependencies with version numbers.
Experiment Setup No The paper is theoretical and does not conduct empirical experiments. Therefore, it does not provide specific experimental setup details such as hyperparameter values or training configurations.