Statistically Significant Concept-based Explanation of Image Classifiers via Model Knockoffs

Authors: Kaiwen Xu, Kazuto Fukuchi, Youhei Akimoto, Jun Sakuma

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

Reproducibility Variable Result LLM Response
Research Type Experimental We evaluate the proposed method in our synthetic and real data experiments.
Researcher Affiliation Academia Kaiwen Xu1,3 , Kazuto Fukuchi1,3 , Youhei Akimoto1,3 and Jun Sakuma2,3 1University of Tsukuba 2Tokyo Institute of Technology 3RIKEN AIP
Pseudocode Yes Algorithm 1 Algorithm for concept selection
Open Source Code No The paper does not include any explicit statement about releasing source code or a link to a code repository.
Open Datasets Yes In our synthetic experiments, we use the Celeb A dataset [Liu et al., 2015]... Colored MNIST. We manually add six types of colors to the MNIST dataset [Le Cun et al., 1998]...
Dataset Splits No The paper states, 'We divided the dataset into two parts, the dataset DL for feature extractor training and the dataset DS for concept selection as shown in step 1.' but does not specify exact percentages, absolute sample counts, or other specific details for training, validation, or test splits. No explicit validation split is mentioned.
Hardware Specification No The paper does not provide any specific hardware details such as GPU/CPU models, memory specifications, or types of computing resources used for experiments.
Software Dependencies No The paper does not provide any specific software dependencies or library versions (e.g., Python, PyTorch, TensorFlow versions) used in the experiments.
Experiment Setup Yes The paper specifies hyperparameter values such as 'α5 = 0.25', 'α5 = 64', 'α5 = 128', and 'α2 = 5' and mentions controlling concept sparsity by adjusting α5.