Risk-Sensitive and Robust Decision-Making: a CVaR Optimization Approach

Authors: Yinlam Chow, Aviv Tamar, Shie Mannor, Marco Pavone

NeurIPS 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Finally, we present results from numerical experiments that corroborate our theoretical findings and show the practicality of our approach.
Researcher Affiliation Academia Yinlam Chow Stanford University ychow@stanford.edu Aviv Tamar UC Berkeley avivt@berkeley.edu Shie Mannor Technion shie@ee.technion.ac.il Marco Pavone Stanford University pavone@stanford.edu
Pseudocode Yes Algorithm 1 CVa R Value Iteration with Linear Interpolation
Open Source Code Yes The Matlab code used for the experiments is provided in the supplementary material.
Open Datasets No The paper describes a custom-generated 'grid-world simulation' for its experiments, rather than using a publicly available or open dataset with concrete access information.
Dataset Splits No The paper describes a simulation environment and evaluation procedure ('trained... on the nominal... evaluated them on 400 perturbed scenarios') but does not provide specific dataset split information (percentages, sample counts, or citations to predefined splits) in the context of training, validation, or test sets.
Hardware Specification No The paper does not provide specific hardware details (exact GPU/CPU models, processor types, or memory amounts) used for running its experiments.
Software Dependencies No The paper mentions 'Matlab code' and 'CPLEX linear programming solver' but does not provide specific version numbers for these software dependencies.
Experiment Setup Yes For our experiments, we choose a 64 53 grid-world (see Figure 1), for a total of 3,312 states... We choose δ = 0.05, and a discount factor γ = 0.95... we set the penalty cost equal to M = 2/(1 γ)... we set ϵ = 0.1 and θ = 2.067