Interpretable Latent Space to Enable Counterfactual Explanations

Bodria, Francesco; Guidotti, Riccardo; Giannotti, Fosca; Pedreschi, Dino

doi:10.1007/978-3-031-18840-4_37

Cited by 7 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In his work, Bodria [10], cites various methods to test the stability of the explainers for black-box models. The method based on the Lipschitz constant, proposed by [20], seemed to us the most appropriate to measure stability of the ExMs in case when the classifier is stable to the noise.…”

Section: State-of-the-art In Evaluation Of Explainers For Cnn Classif...mentioning

confidence: 99%

“…The stability metric for evaluation of black-box explanation methods was proposed in [10]. and is based on the Lipschitz constant.…”

Section: No-reference-based Metric: Stabilitymentioning

confidence: 99%

“…The intuition behind this comparison is that the explainer is good in case of true positives if it shows the area in the image which attracted human attention, as Deep Neural networks are biologically inspired models and are supposed to imitate human brain in decision-making. Other metrics have been recently proposed in [10]. This paper evaluates explanation methods FEM [11], Multilayered FEM (MLFEM) [8], and Grad-CAM [6], using two approaches: the GFDM-based metrics we previously proposed and the so-called stability metric from [10].…”

Section: Introductionmentioning

confidence: 99%

“…Other metrics have been recently proposed in [10]. This paper evaluates explanation methods FEM [11], Multilayered FEM (MLFEM) [8], and Grad-CAM [6], using two approaches: the GFDM-based metrics we previously proposed and the so-called stability metric from [10]. We also study how these metric correlates to open the way to evaluation of the quality of explainers without ground truth.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Evaluation of Explanation Methods of AI - CNNs in Image Classification Tasks with Reference-based and No-reference Metrics

Zhukov¹,

Benois‐Pineau²,

Giot³

2023

AAIML

View full text Add to dashboard Cite

The most popular methods in AI-machine learning paradigm are mainly black boxes. This is why explanation of AI decisions is of emergency. Although dedicated explanation tools have been massively developed, the evaluation of their quality remains an open research question. In this paper, we generalize the methodologies of evaluation of post-hoc explainers of CNNs’ decisions in visual classification tasks with reference and no-reference based metrics. We apply them on our previously developed explainers (FEM1 , MLFEM), and popular Grad-CAM. The reference-based metrics are Pearson correlation coefficient and Similarity computed between the explanation map and its ground truth represented by a Gaze Fixation Density Map obtained with a psycho-visual experiment. As a no-reference metric, we use stability metric, proposed by Alvarez-Melis and Jaakkola. We study its behaviour, consensus with reference-based metrics and show that in case of several kinds of degradation on input images, this metric is in agreement with reference-based ones. Therefore, it can be used for evaluation of the quality of explainers when the ground truth is not available.

show abstract

Section: State-of-the-art In Evaluation Of Explainers For Cnn Classif...mentioning

confidence: 99%

“…The stability metric for evaluation of black-box explanation methods was proposed in [10]. and is based on the Lipschitz constant.…”

Section: No-reference-based Metric: Stabilitymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Evaluation of Explanation Methods of AI - CNNs in Image Classification Tasks with Reference-based and No-reference Metrics

Zhukov¹,

Benois‐Pineau²,

Giot³

2023

AAIML

View full text Add to dashboard Cite

show abstract

“…Следует уточнить, что количество публикаций, так или иначе посвященных проблематике моделей объяснения ИСД, в настоящее время превышает все мыслимые пределы, и их полный обзор не представляется возможным. Например, обзоры, полностью посвященные моделям объяснения в самых различных областях, приведены в [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15]. Обзоры, которые рассматривают только различные аспекты применения моделей объяснения в медицине, приведены в [16][17][18][19][20][21][22][23][24][25][26][27][28].…”

Section: Introductionunclassified

Explanation models as a component of the intelligent computer-aided diagnosis systems in medicine: a brief review

Utkin

Krylova

Konstantinov

2023

jour

View full text Add to dashboard Cite

The paper considers the most important and effective approaches and models for explaining and interpreting diagnostic results obtained using intelligent computer-aided diagnosis systems. The need to use them is due to the fact that the intelligent computer-aided diagnosis system itself is a “black box” and it is important for the doctor not only to get the patient’s diagnosis, but also to understand why such a diagnosis is stated, what elements of the patient information are the most significant from the point of view of the diagnosis. Reviews of the main approaches to explain predictions of machine learning models applied to general areas as well as to medicine are presented. It is shown how different types of the initial patient information impact on the choice of explanation models. Models are considered when visual or tabular information is available. Example-based explanation models are also studied. The purpose of the work is to review the main explanation models and their dependence on types of information about the patient.

show abstract

Machine Learning‐Accelerated Development of Perovskite Optoelectronics Toward Efficient Energy Harvesting and Conversion

Chen,

Huang

2023

Adv Energy and Sustain Res

View full text Add to dashboard Cite

For next‐generation optoelectronic devices with efficient energy harvesting and conversion, designing advanced perovskite materials with exceptional optoelectrical properties is highly critical. However, the conventional trial‐and‐error approaches usually lead to long research periods, high costs, and low efficiency, which hinder the efficient development of optoelectronic devices for broad applications. The machine learning (ML) technique emerges as a powerful tool for materials designs, which supplies promising solutions to break the current bottlenecks in the developments of perovskite optoelectronics. Herein, the fundamental workflow of ML to interpret the working mechanisms step by step from a general perspective is first demonstrated. Then, the significant contributions of ML in designs and explorations of perovskite optoelectronics regarding novel materials discovery, the underlying mechanisms interpretation, and large‐scale information process strategy are illustrated. Based on current research progress, the potential of ML techniques in cross‐disciplinary directions to achieve the boost of material designs and optimizations toward perovskite materials is pointed out. In the end, the current advances of ML in perovskite optoelectronics are summarized and the future development directions are shown. This perspective supplies important insights into the developments of perovskite materials for the next generation of efficient and stable optoelectronic devices.

show abstract

Interpretable Latent Space to Enable Counterfactual Explanations

Cited by 7 publications

References 24 publications

Evaluation of Explanation Methods of AI - CNNs in Image Classification Tasks with Reference-based and No-reference Metrics

Evaluation of Explanation Methods of AI - CNNs in Image Classification Tasks with Reference-based and No-reference Metrics

Explanation models as a component of the intelligent computer-aided diagnosis systems in medicine: a brief review

Machine Learning‐Accelerated Development of Perovskite Optoelectronics Toward Efficient Energy Harvesting and Conversion

Contact Info

Product

Resources

About