Adversarial Training from Mean Field Perspective

Authors: Soichiro Kumano, Hiroshi Kera, Toshihiko Yamasaki

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We validate Thms 5.1, 5.4 and 5.7 via numerical experiments.
Researcher Affiliation Academia Soichiro Kumano The University of Tokyo kumano@cvm.t.u-tokyo.ac.jp Hiroshi Kera Chiba University kera@chiba-u.jp Toshihiko Yamasaki The University of Tokyo yamasaki@cvm.t.u-tokyo.ac.jp
Pseudocode No No pseudocode or algorithm blocks were found in the paper.
Open Source Code No The paper does not provide an explicit statement or link for open-source code for the described methodology.
Open Datasets Yes We used MNIST [26] and Fashion-MNIST [99] as the training dataset.
Dataset Splits No The paper mentions using MNIST and Fashion-MNIST datasets but does not explicitly provide training/test/validation dataset splits or refer to predefined splits with specific details for reproducibility.
Hardware Specification Yes All experiments are conducted on an NVIDIA A100.
Software Dependencies No The paper mentions optimizers like Adam [51] but does not provide specific version numbers for software dependencies such as programming languages or libraries.
Experiment Setup Yes The vanilla Re LU networks were initialized with σ2 w = 2 and σ2 b = 0.01 to meet the (M, m)-trainability conditions (Lemma 5.6). ... We used stochastic gradient descent with a small learning rate, 0.001. ... We set an optimizer to Adam, a learning rate to 0.001, and epochs to 200.