Binary Search with Distributional Predictions

Authors: Michael Dinitz, Sungjin Im, Thomas Lavastida, Ben Moseley, Aidin Niaparast, Sergei Vassilvitskii

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

Reproducibility Variable Result LLM Response
Research Type Experimental We complement this with a lower bound showing that this query complexity is essentially optimal (up to constants), and experiments validating the practical usefulness of our algorithm.
Researcher Affiliation Collaboration Michael Dinitz Johns Hopkins University mdinitz@cs.jhu.edu Sungjin Im UC Merced sim3@ucmerced.edu Thomas Lavastida University of Texas at Dallas thomas.lavastida@utdallas.edu Benjamin Moseley Carnegie Mellon University moseleyb@andrew.cmu.edu Aidin Niaparast Carnegie Mellon University aniapara@andrew.cmu.edu Sergei Vassilvitskii Google Research sergeiv@google.com
Pseudocode No The algorithm is described step-by-step in natural language within Section 3 and Section 4.1, but it is not presented in a formal pseudocode block or labeled 'Algorithm X' figure.
Open Source Code Yes Our implementation can be found at https://github.com/Aidin Niaparast/Learned-BST.
Open Datasets Yes In order to test our approach on real-world data, we use temporal networks from Stanford Large Network Dataset Collection5. These datasets represent the interactions on stack exchange websites Stack Overflow, Ask Ubuntu, and Super User [Paranjape et al., 2017]. 5https://snap.stanford.edu/data/index.html
Dataset Splits Yes For t = 5, 10, . . . , 50, we use the first t percent of A as training data and the rest as test data.
Hardware Specification Yes We use Python 3.10 for conducting our experiments on a system equipped with an 11th Generation Intel Core i7 CPU running at 2.80GHz, 32GB of RAM, a 128GB NVMe KIOXIA disk drive, and a 64-bit Windows 10 Enterprise operating system.
Software Dependencies No The paper mentions 'Python 3.10' as the software used for experiments, but it does not list multiple key software components with their specific version numbers or a self-contained solver/package with a version number.
Experiment Setup Yes We make one modification, setting the parameter d larger to broaden the search space in the very early iterations, setting d to min(2^8 2i, r − ℓ).