Case-Based Reasoning Research and Development : 30th International Conference, ICCBR 2022, Nancy, France, September 12-15, 2022, Proceedings.
Mark T. Keane
Bok Engelsk 2022 · Electronic books.
Omfang | 1 online resource (420 pages)
|
---|---|
Opplysninger | Intro -- Preface -- Organization -- Invited Talks -- Seeing Through Black Boxes with Human Vision: Deep Learning and Explainable AI in Medical Image Applications -- Case-Based Reasoning for Clinical Decisions That Are Computer-Aided, Not Automated -- Towards More Cognitively Appealing Paradigms in Case-Based Reasoning -- Contents -- Explainability in CBR -- Using Case-Based Reasoning for Capturing Expert Knowledge on Explanation Methods -- 1 Introduction -- 2 Background -- 3 Case-Based Elicitation -- 3.1 Case Structure -- 3.2 Case Base Acquisition -- 4 CBR Process -- 5 Evaluation and Discussion -- 6 Conclusions -- References -- A Few Good Counterfactuals: Generating Interpretable, Plausible and Diverse Counterfactual Explanations -- 1 Introduction -- 2 Related Work -- 2.1 What Are Good Counterfactual Explanations? -- 2.2 Perturbation-Based Approaches -- 2.3 Instance-Based Approaches -- 2.4 Instance-Based Shortcomings -- 3 Good Counterfactuals in Multi-class Domains -- 3.1 Reusing the kNN Explanation Cases -- 3.2 Validating Candidate Counterfactuals -- 3.3 Discussion -- 4 Evaluation -- 4.1 Methodology -- 4.2 Results -- 5 Conclusions -- References -- How Close Is Too Close? The Role of Feature Attributions in Discovering Counterfactual Explanations -- 1 Introduction -- 2 Related Work -- 3 DisCERN -- 3.1 Nearest-Unlike Neighbour -- 3.2 Feature Ordering by Feature Attribution -- 3.3 Substitution-Based Adaptation -- 3.4 Integrated Gradients for DisCERN -- 3.5 Bringing the NUN Closer -- 4 Evaluation -- 4.1 Datasets -- 4.2 Experiment Setup -- 4.3 Performance Measures for Counterfactual Explanations -- 5 Results -- 5.1 A Comparison of Feature Attribution Techniques -- 5.2 A Comparison of Counterfactual Discovery Algorithms -- 5.3 Impact of Bringing NUN Closer -- 6 Conclusions -- References -- Algorithmic Bias and Fairness in Case-Based Reasoning.. - 1 Introduction -- 2 Related Research -- 2.1 Bias in ML -- 2.2 Bias in CBR -- 2.3 Metric Learning -- 3 FairRet: Eliminating Bias with Metric Learning -- 3.1 Bias and The Similarity Knowledge Container -- 3.2 A Metric Learning Approach -- 3.3 Multi-objective Particle Swarm Optimization -- 4 Results -- 4.1 Dealing with Underestimation Bias -- 4.2 Outcome Distortion -- 4.3 Retrieval Overlap -- 5 Conclusions -- References -- "Better" Counterfactuals, Ones People Can Understand: Psychologically-Plausible Case-Based Counterfactuals Using Categorical Features for Explainable AI (XAI) -- 1 Introduction -- 2 Background: Computation and Psychology of Counterfactuals -- 2.1 User Studies of Counterfactual XAI: Mixed Results -- 3 Study 1: Plotting Counterfactuals that have Categoricals -- 3.1 Results and Discussion -- 4 Transforming Case-Based Counterfactuals, Categorically -- 4.1 Case-Based Counterfactual Methods: CB1-CF and CB2-CF -- 4.2 Counterfactuals with Categorical Transforms #1: Global Binning -- 4.3 Counterfactuals with Categorical Transforms #2: Local Direction -- 5 Study 2: Evaluating CAT-CF Methods -- 5.1 Method: Data and Procedure -- 5.2 Results and Discussion: Counterfactual Distance -- 6 Conclusions -- References -- Representation and Similarity -- Extracting Case Indices from Convolutional Neural Networks: A Comparative Study -- 1 Introduction -- 2 Potential Feature Extraction Points in cnns -- 3 Related Work -- 4 Three Structure-Based Feature Extraction Methods -- 4.1 Post-convolution Feature Extraction -- 4.2 Post-dense Feature Extraction -- 4.3 Multi-net Feature Extraction -- 5 Evaluation -- 5.1 Hypotheses -- 5.2 Test Domain and Test Set Selection -- 5.3 Testbed System -- 5.4 Accuracy Testing and Informal Upper Bound -- 6 Results and Discussion -- 6.1 Comparative Performance -- 6.2 Discussion -- 7 Ramifications for Interpretability.. - 3 Boundary Graphs and Labeled Boundary Multigraphs -- 3.1 Boundary Graphs -- 3.2 Labeled Boundary Multigraphs -- 3.3 Discussion -- 4 Empirical Evaluation -- 4.1 Experimental Set-Up -- 4.2 Classical Benchmark Data Sets -- 4.3 Scaling Analysis -- 5 Conclusion -- References -- Particle Swarm Optimization in Small Case Bases for Software Effort Estimation -- 1 Introduction -- 2 Related Work -- 3 Software Effort Estimation of User Stories -- 4 CBR Approach -- 4.1 Case Representation -- 4.2 Similarity -- 4.3 Adaptation -- 4.4 Weight Optimization with PSO -- 5 Experiments -- 5.1 Experimental Data -- 5.2 Experiment 1 -- 5.3 Experiment 2 -- 5.4 Discussion of Results -- 6 Conclusion -- References -- MicroCBR: Case-Based Reasoning on Spatio-temporal Fault Knowledge Graph for Microservices Troubleshooting -- 1 Introduction -- 2 Related Work -- 3 Background and Motivation -- 3.1 Background with Basic Concepts -- 3.2 Motivation -- 4 Troubleshooting Framework -- 4.1 Framework Overview -- 4.2 Spatio-Temporal Fault Knowledge Graph -- 4.3 Fingerprinting the Fault -- 4.4 Case-Based Reasoning -- 5 Evaluation -- 5.1 Evaluation Setup -- 5.2 Q1. Comparative Experiments -- 5.3 Q2. Ablation Experiment -- 5.4 Q3. Efficiency Experiments -- 5.5 Q4. Case Studies and Learned Lessons -- 6 Conclusion -- References -- .26em plus .1em minus .1emGPU-Based Graph Matching for Accelerating Similarity Assessment in Process-Oriented Case-Based Reasoning -- 1 Introduction -- 2 Foundations and Related Work -- 2.1 Semantic Workflow Graph Representation -- 2.2 State-Space Search by Using A* -- 2.3 Related Work -- 3 AMonG: A*-Based Graph Matching on Graphic Processing Units -- 3.1 Overview and Components -- 3.2 Parallel Graph Matching -- 4 Experimental Evaluation -- 4.1 Experimental Setup -- 4.2 Experimental Results -- 4.3 Discussion and Further Considerations -- 5 Conclusion and Future Work.. - 6 Conclusion -- References -- Adaptation and Analogical Reasoning -- Case Adaptation with Neural Networks: Capabilities and Limitations -- 1 Introduction -- 2 Background -- 3 NN-CDH for both Classification and Regression -- 3.1 General Model of Case Adaptation -- 3.2 1-Hot/1-Cold Nominal Difference -- 3.3 Neural Network Structure of NN-CDH -- 3.4 Training and Adaptation Procedure -- 4 Evaluation -- 4.1 Systems Being Compared -- 4.2 Assembling Case Pairs for Training -- 4.3 Data Sets -- 4.4 Artificial Data Sets -- 5 Conclusion -- References -- A Deep Learning Approach to Solving Morphological Analogies -- 1 Introduction -- 2 The Problem of Morphological Analogy -- 3 Proposed Approach -- 3.1 Classification, Retrieval and Embedding Models -- 3.2 Training and Evaluation -- 4 Experiments -- 4.1 Data -- 4.2 Refining the Training Procedure -- 4.3 Performance Comparison with State of the Art Methods -- 4.4 Distance of the Expected Result -- 4.5 Case Analysis: Navajo and Georgian -- 5 Conclusion and Perspectives -- References -- Theoretical and Experimental Study of a Complexity Measure for Analogical Transfer -- 1 Introduction -- 2 Reminder on Complexity-Based Analogy -- 2.1 Notations -- 2.2 Ordinal Analogical Principle: Complexity Definition -- 2.3 Ordinal Analogical Inference Algorithm -- 3 Theoretical Property of the Complexity Measure: Upper Bound -- 3.1 General Case -- 3.2 Binary Classification Case -- 4 Algorithmic Optimisation -- 4.1 Principle -- 4.2 Proposed Optimized Algorithm -- 5 Experimental Study -- 5.1 Computational Cost -- 5.2 Correlation Between Case Base Complexity and Performance -- 5.3 Correlation Between Complexity and Task Difficulty -- 6 Conclusion and Future Works -- References -- Graphs and Optimisation -- Case-Based Learning and Reasoning Using Layered Boundary Multigraphs -- 1 Introduction -- 2 Background and Related Work.. - 8 Conclusions and Future Work -- References -- Exploring the Effect of Recipe Representation on Critique-Based Conversational Recommendation -- 1 Introduction -- 2 Background -- 2.1 Diversity in Recommender Systems -- 2.2 Critique-Based Conversational Recommender Systems -- 2.3 Diversity in Recipe Recommenders -- 3 DiversityBite Framework: Recommend, Review, Revise -- 3.1 Adaptive Diversity Goal Approach -- 4 Evaluation -- 4.1 Case Base -- 4.2 Implementation: DGF, AGD, and Diversity Scoring -- 4.3 Simulation Study: Incorporating Diversity in Critique -- 4.4 User Study: Comparing Different Recipe Representations -- 5 Conclusion -- References -- Explaining CBR Systems Through Retrieval and Similarity Measure Visualizations: A Case Study -- 1 Introduction -- 2 Related Work -- 3 SupportPrim CBR System -- 3.1 Data -- 3.2 Case Representation and Similarity Modeling -- 3.3 Case Base and Similarity Population -- 4 Explanatory Case Base Visualizations -- 4.1 Accessing the CBR System's Model -- 4.2 Visualization of Retrievals -- 4.3 Visualization of the Similarity Scores for Individual Case Comparisons -- 5 Experiments -- 6 Discussion -- 7 Conclusion -- References -- Adapting Semantic Similarity Methods for Case-Based Reasoning in the Cloud -- 1 Introduction -- 2 Related Work -- 2.1 Clood CBR -- 2.2 Ontologies in CBR -- 2.3 Retrieval with Word Embedding -- 2.4 Serverless Function Benefits and Limitations -- 3 Semantic Similarity Metrics in a Microservices Architecture -- 3.1 Clood Similarity Functions Overview -- 3.2 Similarity Table -- 3.3 Word Embedding Based Similarity -- 3.4 Ontology-Based Similarity Measure -- 4 Implementation of Semantic Similarity Measures on Clood Framework -- 4.1 Word Embedding Similarity on Clood -- 4.2 Ontology-Based Similarity on Clood -- 5 Evaluation of Resource Impact -- 5.1 Experiment Setup -- 5.2 Result and Discussion.. - References.
|
Sjanger | |
ISBN | 9783031149238
|
ISBN(galt) |