Explanation-Guided Training for Cross-Domain Few-Shot Classification

Sun, Jiamei; Lapuschkin, Sebastian; Samek, Wojciech; Zhao, Yunqing; Cheung, Ngai-Man; Binder, Alexander

doi:10.48550/arxiv.2007.08790

Cited by 8 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Motivated by it, the researchers [22] used LRP in GNN to obtain insights into the black-box of GNN models. Schnake et al [17] proposed GNN-LRP based higher-order Taylor expansions.…”

Section: Explaining Graph Neural Network Through Xai Methodsmentioning

confidence: 99%

A Survey of Explainable Graph Neural Networks: Taxonomy and Evaluation Metrics

Li¹,

Zhou²,

Verma³

et al. 2022

Preprint

View full text Add to dashboard Cite

Graph neural networks (GNNs) have demonstrated a significant boost in prediction performance on the graph data. At the same time, the predictions made by these models are often hard to interpret. In that regard, many efforts have been made to explain the prediction mechanisms of these models from perspectives such as GNNExplainer, XGNN and PGExplainer. Although such works present systematic frameworks to interpret GNNs, a holistic review for explainable GNNs is unavailable. In this survey, we present a comprehensive review of explainability techniques developed for GNNs. We focus on explainable graph neural networks, categorize them based on the use of explainable methods. We further provide the common performance metrics for GNNs explanations and point out several future research directions.

show abstract

“…Motivated by it, the researchers [22] used LRP in GNN to obtain insights into the black-box of GNN models. Schnake et al [17] proposed GNN-LRP based higher-order Taylor expansions.…”

Section: Explaining Graph Neural Network Through Xai Methodsmentioning

confidence: 99%

A Survey of Explainable Graph Neural Networks: Taxonomy and Evaluation Metrics

Li¹,

Zhou²,

Verma³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…• We introduce a novel method of explanation-guided training of DNN to prioritize relevant features in the input during the training phase. • While the authors in the related work [28] presented a method of explanation-guided training to dynamically finds and emphasize the features which are important, our method aims to allow domain experts to prioritize important features in the input image by segmentation mask. Hence, the proposed method is useful when we want to prefer some well-known input features, but we do not want to restrict the model only to these features.…”

Section: Our Contributionmentioning

confidence: 99%

Improving a neural network model by explanation-guided training for glioma classification based on MRI data

Sefcik

Benešová

2023

Int. j. inf. tecnol.

View full text Add to dashboard Cite

In recent years, artificial intelligence systems have come to the forefront. These systems, mostly based on deep learning, achieve excellent results in areas such as image processing, natural language processing and speech recognition. Despite the statistically high accuracy of deep learning models, their output is often based on ”black box” decisions. Thus, interpretability methods (Reyes et al. in Radiol Artif Intell 2(3):e190043, 2020) have become a popular way to gain insight into the decision-making process of deep learning models (Miller in Artif Intell 267:1–38, 2019). Explanation of deep learning models is desirable in the medical domain since experts have to justify their judgments to the patients. In this work, we proposed a method for explanation-guided training that uses a layer-wise relevance propagation technique to force the model to focus only on the relevant part of the image. We experimentally verified our method on a convolutional neural network model for low-grade and high-grade glioma classification problems. Our experiments produced promising results in the way where we use interpretation techniques in the training process.

show abstract

“…Meta-learning performance suffers on VTAB-v2 In contrast to BiT, Figure 1 shows that meta-learning approaches struggle to compete with transfer learning on VTAB-v2. MD- tunity to apply existing cross-domain few-shot classification approaches (Tseng et al, 2020;Sun et al, 2020;Phoo & Hariharan, 2020;Liu et al, 2020;Cai & Shen, 2020) at scale.…”

Section: Comparison Of Selected Approachesmentioning

confidence: 99%

Comparing Transfer and Meta Learning Approaches on a Unified Few-Shot Classification Benchmark

Dumoulin¹,

Houlsby²,

Evci³

et al. 2021

Preprint

View full text Add to dashboard Cite

Meta and transfer learning are two successful families of approaches to few-shot learning. Despite highly related goals, state-of-the-art advances in each family are measured largely in isolation of each other. As a result of diverging evaluation norms, a direct or thorough comparison of different approaches is challenging. To bridge this gap, we perform a cross-family study of the best transfer and meta learners on both a large-scale meta-learning benchmark (Meta-Dataset, MD), and a transfer learning benchmark (Visual Task Adaptation Benchmark, VTAB). We find that, on average, large-scale transfer methods (Big Transfer, BiT) outperform competing approaches on MD, even when trained only on ImageNet. In contrast, meta-learning approaches struggle to compete on VTAB when trained and validated on MD. However, BiT is not without limitations, and pushing for scale does not improve performance on highly out-of-distribution MD tasks. In performing this study, we reveal a number of discrepancies in evaluation norms and study some of these in light of the performance gap. We hope that this work facilitates sharing of insights from each community, and accelerates progress on fewshot learning.

show abstract

Explanation-Guided Training for Cross-Domain Few-Shot Classification

Cited by 8 publications

References 36 publications

A Survey of Explainable Graph Neural Networks: Taxonomy and Evaluation Metrics

A Survey of Explainable Graph Neural Networks: Taxonomy and Evaluation Metrics

Improving a neural network model by explanation-guided training for glioma classification based on MRI data

Comparing Transfer and Meta Learning Approaches on a Unified Few-Shot Classification Benchmark

Contact Info

Product

Resources

About