Inverse Reinforcement Learning From Like-Minded Teachers

Authors: Ritesh Noothigattu, Tom Yan, Ariel D. Procaccia9197-9204

AAAI 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We next study the empirical performance of our algorithm for the inverse multi-armed bandit problem.
Researcher Affiliation Academia 1 Carnegie Mellon University 2 Harvard University {riteshn, tyyan}@cmu.edu, arielpro@seas.harvard.edu
Pseudocode No For completeness we present these algorithms, and formally state their featurematching guarantees, in the full version of the paper.
Open Source Code No The paper does not contain any explicit statements or links indicating the availability of open-source code for the described methodology.
Open Datasets No In the first set of experiments, we fix the noise standard deviation σ to 1, generate n = 500 agents according to the noise η N(0, σ2), and vary parameter δ from 0.01 to 3.
Dataset Splits No The paper describes generating data for simulations ('generate n = 500 agents') and averaging results over multiple runs ('averaged over 1000 runs'), but it does not specify traditional training/validation/test dataset splits for a fixed dataset.
Hardware Specification No The paper does not provide specific details about the hardware used to run the experiments.
Software Dependencies No The paper does not specify any software dependencies with version numbers.
Experiment Setup Yes In the first set of experiments, we fix the noise standard deviation σ to 1, generate n = 500 agents according to the noise η N(0, σ2), and vary parameter δ from 0.01 to 3. ... Next, we fix the parameter δ to 1 and generate n = 500 agents according to noise η N(0, σ2), while varying the noise parameter σ from 0.01 to 5.