More Like this

1. Deep Descriptive Clustering

   Zhang, H; Davidson, I (August 2021, IJCAI)

   null (Ed.)

   Recent work on explainable clustering allows describing clusters when the features are interpretable. However, much modern machine learning focuses on complex data such as images, text, and graphs where deep learning is used but the raw features of data are not interpretable. This paper explores a novel setting for performing clustering on complex data while simultaneously generating explanations using interpretable tags. We propose deep descriptive clustering that performs sub-symbolic representation learning on complex data while generating explanations based on symbolic data. We form good clusters by maximizing the mutual information between empirical distribution on the inputs and the induced clustering labels for clustering objectives. We generate explanations by solving an integer linear programming that generates concise and orthogonal descriptions for each cluster. Finally, we allow the explanation to inform better clustering by proposing a novel pairwise loss with self-generated constraints to maximize the clustering and explanation module's consistency. Experimental results on public data demonstrate that our model outperforms competitive baselines in clustering performance while offering high-quality cluster-level explanations. 

   more » « less
2. Concept-Centric Transformers: Enhancing Model Interpretability through Object-Centric Concept Learning within a Shared Global Workspace

   https://doi.org/10.1109/WACV57701.2024.00481  

   Hong, Jinyung; Park, Keun Hee; Pavlic, Theodore P (January 2024, IEEE)

   Many interpretable AI approaches have been proposed to provide plausible explanations for a model’s decision-making. However, configuring an explainable model that effectively communicates among computational modules has received less attention. A recently proposed shared global workspace theory showed that networks of distributed modules can benefit from sharing information with a bottle-necked memory because the communication constraints encourage specialization, compositionality, and synchronization among the modules. Inspired by this, we propose Concept-Centric Transformers, a simple yet effective configuration of the shared global workspace for interpretability, consisting of: i) an object-centric-based memory module for extracting semantic concepts from input features, ii) a cross-attention mechanism between the learned concept and input embeddings, and iii) standard classification and explanation losses to allow human analysts to directly assess an explanation for the model’s classification reasoning. We test our approach against other existing concept-based methods on classification tasks for various datasets, including CIFAR100, CUB-200-2011, and ImageNet, and we show that our model achieves better classification accuracy than all baselines across all problems but also generates more consistent concept-based explanations of classification output. 

   more » « less
3. Interpretable Active Learning

   Phillips, Richard; Chang, Kyu Hyun; Friedler, Sorelle A. (January 2018, Conference on Fairness, Accountability, and Transparency)

   Active learning has long been a topic of study in machine learning. However, as increasingly complex and opaque models have become standard practice, the process of active learning, too, has become more opaque. There has been little investigation into interpreting what specific trends and patterns an active learning strategy may be exploring. This work expands on the Local Interpretable Model-agnostic Explanations framework (LIME) to provide explanations for active learning recommendations. We demonstrate how LIME can be used to generate locally faithful explanations for an active learning strategy, and how these explanations can be used to understand how different models and datasets explore a problem space over time. These explanations can also be used to generate batches based on common sources of uncertainty. These regions of common uncertainty can be useful for understanding a model’s current weaknesses. In order to quantify the per-subgroup differences in how an active learning strategy queries spatial regions, we introduce a notion of uncertainty bias (based on disparate impact) to measure the discrepancy in the confidence for a model’s predictions between one subgroup and another. Using the uncertainty bias measure, we show that our query explanations accurately reflect the subgroup focus of the active learning queries, allowing for an interpretable explanation of what is being learned as points with similar sources of uncertainty have their uncertainty bias resolved. We demonstrate that this technique can be applied to track uncertainty bias over user-defined clusters or automatically generated clusters based on the source of uncertainty. We also measure how the choice of initial labeled examples effects groups over time. 

   more » « less
4. Interpretable Active Learning

   Phillips, Richard; Chang, Kyu Hyun; Friedler, Sorelle A. (January 2018, Conference on Fairness, Accountability, and Transparency)

   Active learning has long been a topic of study in machine learning. However, as increasingly complex and opaque models have become standard practice, the process of active learning, too, has become more opaque. There has been little investigation into interpreting what specific trends and patterns an active learning strategy may be exploring. This work expands on the Local Interpretable Model-agnostic Explanations framework (LIME) to provide explanations for active learning recommendations. We demonstrate how LIME can be used to generate locally faithful explanations for an active learning strategy, and how these explanations can be used to understand how different models and datasets explore a problem space over time. In order to quantify the per-subgroup differences in how an active learning strategy queries spatial regions, we introduce a notion of uncertainty bias (based on disparate impact) to measure the discrepancy in the confidence for a model's predictions between one subgroup and another. Using the uncertainty bias measure, we show that our query explanations accurately reflect the subgroup focus of the active learning queries, allowing for an interpretable explanation of what is being learned as points with similar sources of uncertainty have their uncertainty bias resolved. We demonstrate that this technique can be applied to track uncertainty bias over user-defined clusters or automatically generated clusters based on the source of uncertainty. 

   more » « less
5. Constructive Interpretability with CoLabel: Corroborative Integration, Complementary Features, and Collaborative Learning

   https://doi.org/10.1109/CogMI56440.2022.00021  

   Suprem, Abhijit; Vaidya, Sanjyot; Cherkadi, Suma; Singh, Purva; Ferreira, Joao Eduardo; Pu, Calton (December 2022, Proceedings of 2022 IEEE International Conference on Cognitive Machine Intelligence (CogMI))

   Machine learning models with explainable predictions are increasingly sought after, especially for real-world, mission-critical applications that require bias detection and risk mitigation. Inherent interpretability, where a model is designed from the ground-up for interpretability, provides intuitive insights and transparent explanations on model prediction and performance. In this paper, we present COLABEL, an approach to build interpretable models with explanations rooted in the ground truth. We demonstrate COLABEL in a vehicle feature extraction application in the context of vehicle make-model recognition (VMMR). By construction, COLABEL performs VMMR with a composite of interpretable features such as vehicle color, type, and make, all based on interpretable annotations of the ground truth labels. First, COLABEL performs corroborative integration to join multiple datasets that each have a subset of desired annotations of color, type, and make. Then, COLABEL uses decomposable branches to extract complementary features corresponding to desired annotations. Finally, COLABEL fuses them together for final predictions. During feature fusion, COLABEL harmonizes complementary branches so that VMMR features are compatible with each other and can be projected to the same semantic space for classification. With inherent interpretability, COLABEL achieves superior performance to the state-of-the-art black-box models, with accuracy of 0.98, 0.95, and 0.94 on CompCars, Cars196, and BoxCars116K, respectively. COLABEL provides intuitive explanations due to constructive interpretability, and subsequently achieves high accuracy and usability in mission-critical situations. 

   more » « less