Heterogeneous Multi-Agent Bandits with Parsimonious Hints

Authors: Amirmahdi Mirfakhar, Xuchuang Wang, Jinhang Zuo, Yair Zick, Mohammad Hajiesmaili

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

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we establish lower bounds to prove the optimality of our results and verify them through numerical simulations. We executed the algorithms HCLA, GP-HCLA, G-HCLA, HD-ETC, and EBHD-ETC with M = 4, K = 4, and match min 0.18, averaging regret and hint complexity over 50 replications for 105 rounds.
Researcher Affiliation Academia 1University of Massachusetts Amherst, 2City University of Hong Kong EMAIL, EMAIL, EMAIL, EMAIL, EMAIL
Pseudocode Yes Algorithm 1: Hinted Centralized Learning Algorithm (HCLA) Algorithm 2: Generalized Projection-based Hinted Centralized Learning Algorithm (GP-HCLA) Algorithm 3: Hinted Decentralized Explore then Commit (HD-ETC) : agent m Algorithm 4: Elimination-Based Hinted Decentralized Explore then Commit (EBHD-ETC) : agent m
Open Source Code No The paper does not provide any explicit statement or link regarding the availability of source code for the described methodology.
Open Datasets No We executed the algorithms HCLA, GP-HCLA, G-HCLA, HD-ETC, and EBHD-ETC with M = 4, K = 4, and match min 0.18, averaging regret and hint complexity over 50 replications for 105 rounds. The experiments appear to be based on simulated environments rather than a specific external dataset, and no dataset is mentioned as publicly available.
Dataset Splits No The paper does not use an external dataset, instead, it describes numerical simulations based on specified parameters (M, K, match min), so there are no dataset splits to provide.
Hardware Specification No The paper mentions numerical simulations and experiments, but does not provide any specific details about the hardware used to run these experiments.
Software Dependencies No The paper describes algorithms and numerical simulations but does not specify any software dependencies or their version numbers used for implementation or experimentation.
Experiment Setup Yes We executed the algorithms HCLA, GP-HCLA, G-HCLA, HD-ETC, and EBHD-ETC with M = 4, K = 4, and match min 0.18, averaging regret and hint complexity over 50 replications for 105 rounds.