Exploitability Minimization in Games and Beyond
Authors: Denizalp Goktas, Amy Greenwald
NeurIPS 2022 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this section, we report on experiments that demonstrate the e ectiveness of our algorithms as compared to others that were also designed to compute GNE. We present numerical results on benchmark psuedo-games that were studied empirically in previous work [104]: 1) monotone pseudo-games with a ne constraints and 2) bilinear pseudo-games with jointly convex constraints.6 We compare our algorithms to the accelerated mirror-prox quadratic penalty method (AMPQP), the only GNEnding algorithm with theoretical convergence guarantees and rates [78].7 |
| Researcher Affiliation | Academia | Denizalp Goktas Department of Computer Science Brown University Providence, RI 02906, USA denizalp_goktas@brown.edu Amy Greenwald Brown University Providence, RI 02906, USA amy_greenwald@brown.edu |
| Pseudocode | Yes | Algorithm 1 Extragradient descent ascent (EDA) Inputs: n, u, X, , T, a(0), b(0) Outputs: (a(t), b(t))T t=0 1: for t = 0, . . . , T 1 do 2: a(t+1/2) = X a(t) ra (a(t), b(t)) 3: b(t+1/2) = X b(t) + rb (a(t), b(t)) 4: a(t+1) = X a(t) ra (a(t+1/2), b(t+1/2)) 5: b(t+1) = X b(t) + rb (a(t+1/2), b(t+1/2)) 6: return (a(t), b(t))T |
| Open Source Code | Yes | All our code can be found on https://github.com/denizalp/exploit-min. |
| Open Datasets | No | The paper uses benchmark pseudo-games and mentions "monotone pseudo-games with a ne constraints" and "bilinear pseudo-games with jointly convex constraints" as studied in previous work [104], but does not provide specific access information (link, citation with authors/year) for the datasets used in their experiments. It describes the characteristics of the games rather than a specific dataset. |
| Dataset Splits | No | The paper does not explicitly mention training, validation, or test dataset splits. It describes the types of pseudo-games used for experiments but not how the data within these games was partitioned for training, validation, or testing. |
| Hardware Specification | No | Appendix A mentions "The experiments were run on a single machine with a 2.3 GHz 8-Core Intel Core i9 processor, 16 GB of DDR4 RAM, and a Radeon Pro 5500M 8 GB GPU." This information is in the appendix, not the main paper. The question specifically asks for hardware details in the main text. |
| Software Dependencies | No | The paper lists general software used in Appendix A (Python [106], NumPy [107], CVXPY [108], Matplotlib [109]) but does not provide specific version numbers for these dependencies within the main text. |
| Experiment Setup | Yes | In the monotone pseudo-games, for EDA we use a constant learning rate of = 0.02, while for ADA we use the constant learning rates of a = 0.02 and a = 0.05. These rates approximate the Lipschitz-smoothness parameter of the pseudo-games. In the bilinear pseudo-games, we use the theoretically-grounded (Theorem 3.4) learning rate of = 1/k Qk for EDA, and a = k Qk + k Qk2 and b = 1/k Qk for ADA. In both settings, we use a regularization of = 0.1 for ADA, which we found by grid search. Likewise, in our implementation of AMPQP, grid search led us to initialize β0 = 0.01, γ = 1.05, = 0.2 in both settings. Additionally, because the iterates generated by AMPQP are not necessarily feasible, we projected them onto the feasible set of actions, so as to avoid in nite exploitability. This heuristic improved the convergence rate of AMPQP signi cantly. The rst iterate for all methods was common, and initialized at random. |