Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Adversarial Fisher Vectors for Unsupervised Representation Learning
Authors: Shuangfei Zhai, Walter Talbott, Carlos Guestrin, Joshua Susskind
NeurIPS 2019 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We conduct experiments showing that the GAN-induced Fisher Vectors demonstrate competitive performance as unsupervised feature extractors for classification and perceptual similarity tasks. Code is available at https://github.com/apple/ml-afv. |
| Researcher Affiliation | Industry | Shuangfei Zhai Walter Talbott Carlos Guestrin Joshua M. Susskind Apple Inc. EMAIL |
| Pseudocode | No | The paper does not contain any structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Code is available at https://github.com/apple/ml-afv. |
| Open Datasets | Yes | We conduct our experiments on images of size 32 × 32, using CIFAR10 and CIFAR100 [33], Celeb A [34] and Image Net [35]. |
| Dataset Splits | No | The paper refers to 'validation examples' and 'validation Fisher Similarity' but does not provide specific percentages or counts for training, validation, or test dataset splits. |
| Hardware Specification | No | The paper does not provide specific hardware details such as GPU models, CPU types, or memory used for running the experiments. |
| Software Dependencies | No | The paper mentions optimizers and normalization techniques, but it does not provide specific version numbers for any software dependencies, libraries, or frameworks used in the experiments. |
| Experiment Setup | Yes | Unless otherwise mentioned, the channel size for convolutional layers is 128. All experiments use batch size 64, ADAM optimizer [37] with β1 = 0, β2 = .999, learning rate for G = 2 × 10−4, and learning rate for D = 4 × 10−4. By default we train our model with a fixed number of iterations (800K) and obtain the last checkpoint for evaluation, unless otherwise mentioned. |