How to Boost Any Loss Function

Authors: Richard Nock, Yishay Mansour

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical Our paper is a theory paper: all claims are properly formalized and used.
Researcher Affiliation Collaboration Richard Nock Google Research richardnock@google.com Yishay Mansour Tel Aviv University Google Research mansour@google.com
Pseudocode Yes Algorithm 1 SECBOOST(S, T) ... Algorithm 2 SOLVEα(S, w, h) ... Algorithm 3 SOLVE_extended(S, w, h, M) ... Algorithm 4 OO_simple(F, et, et 1, z, Z)
Open Source Code No Does the paper provide open access to the data and code, with sufficient instructions to faithfully reproduce the main experimental results, as described in supplemental material? Answer: [No] Justification: Our paper is a theory paper. All algorithms we introduce are either in the main file or the appendix.
Open Datasets Yes We provide an experiment on public domain UCI tictactoe [23] (using a 10-fold stratified crossvalidation to estimate test errors).
Dataset Splits Yes We provide an experiment on public domain UCI tictactoe [23] (using a 10-fold stratified crossvalidation to estimate test errors).
Hardware Specification No For each experiment, does the paper provide sufficient information on the computer resources (type of compute workers, memory, time of execution) needed to reproduce the experiments? Answer: [NA] . Justification: Our paper is a theory paper.
Software Dependencies No Does the paper provide SPECIFIC ANCILLARY SOFTWARE DETAILS (e.g., library or solver names with version numbers like Python 3.8, CPLEX 12.4) needed to replicate the experiment? Answer: [NA] Justification: Our paper is a theory paper.
Experiment Setup Yes the size of the trees (either they have a single internal node = stumps or at most 20 nodes) and, to give one example of how changing a (key) hyperparameter can change the result, we have tested for a scale of changes on the initial value of δ in (60). ... δ 0.1 δ 1.0 ... We flip each label in the training sample with probability η.