InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
Authors: Xi Chen, Yan Duan, Rein Houthooft, John Schulman, Ilya Sutskever, Pieter Abbeel
NeurIPS 2016 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | To evaluate whether the mutual information between latent codes c and generated images G(z, c) can be maximized efficiently with proposed method, we train Info GAN on MNIST dataset with a uniform categorical distribution on latent codes c Cat(K = 10, p = 0.1). In Fig 1, the lower bound LI(G, Q) is quickly maximized to H(c) 2.30, which means the bound (4) is tight and maximal mutual information is achieved. |
| Researcher Affiliation | Collaboration | Xi Chen , Yan Duan , Rein Houthooft , John Schulman , Ilya Sutskever , Pieter Abbeel UC Berkeley, Department of Electrical Engineering and Computer Sciences Open AI |
| Pseudocode | No | The paper describes the algorithm using text and mathematical formulations but does not include a clearly labeled pseudocode block or algorithm. |
| Open Source Code | No | The paper provides a link to an up-to-date version of the paper on arXiv, but not to any source code for the methodology. |
| Open Datasets | Yes | Specifically, Info GAN successfully disentangles writing styles from digit shapes on the MNIST dataset, pose from lighting of 3D rendered images, and background digits from the central digit on the SVHN dataset. It also discovers visual concepts that include hair styles, presence/absence of eyeglasses, and emotions on the Celeb A face dataset. |
| Dataset Splits | No | Detailed experimental setup is described in Appendix. (Appendix not provided in the given text, thus specific dataset split information is not available in the provided content). |
| Hardware Specification | No | The paper does not provide specific hardware details (e.g., GPU/CPU models, memory amounts) used for running its experiments. |
| Software Dependencies | No | The paper mentions using techniques from DC-GAN [16] but does not specify any software names with version numbers (e.g., Python, PyTorch, TensorFlow versions). |
| Experiment Setup | Yes | Even though Info GAN introduces an extra hyperparameter λ, it s easy to tune and simply setting to 1 is sufficient for discrete latent codes. When the latent code contains continuous variables, a smaller λ is typically used to ensure that λLI(G, Q), which now involves differential entropy, is on the same scale as GAN objectives. |