Search

Recent advancements in developing pre-trained models using large-scale datasets have emphasized the importance of robust protocols to adapt them effectively to domain-specific data, especially when the available data is limited. To achieve data-efficient fine-tuning of pre-trained object detection models, data augmentations are crucial. However, selecting the appropriate augmentation policy for a given dataset is known to be challenging. In this study, we address an overlooked aspect of this problem - can bounding box annotations be utilized to develop more effective augmentation policies? Our approach InterAug reveals that, by leveraging the annotations, one can deduce the optimal context for each object in a scene, rather than manipulating the entire scene or just the pre-defined bounding boxes. Through rigorous empirical research involving various benchmarks and architectures, we showcase the effectiveness of InterAug in enhancing robustness, handling data scarcity, and maintaining resilience to diverse high background contexts. An important advantage of InterAug is its compatibility with any off-the-shelf policy, requiring no modifications to the model architecture, and it significantly outperforms existing protocols. We will release the codes upon acceptance. Our codes can be found at https://github.com/kowshikthopalli/InterAug.

Unsupervised domain adaptation (UDA) aims to transfer and adapt knowledge from a labeled source domain to an unlabeled target domain. Traditionally, geometry-based alignment methods, e.g., Orthogonal Procrustes Alignment (OPA), formed an important class of solutions to this problem. Despite their mathematical tractability, they rarely produce effective adaptation performance with the recent benchmarks. Instead, state-of-the-art approaches rely on sophisticated distribution alignment strategies such as adversarial training. In this paper, we show that, conventional OPA, when coupled with powerful deep feature extractors and a novel bi-level optimization formulation, is indeed an effective choice for handling challenging distribution shifts. When compared to existing UDA methods, our approach offers the following benefits: computational efficiency: Through the isolation of alignment and classifier training steps during adaptation, and the use of deep OPA, our approach is computationally very effective (typically requiring only 700 K parameters more than the base feature extractor as compared to millions of extra parameters required by state-of-the-art UDA baselines); (ii) data efficiency: Our approach does not require updating our feature extractor during adaptation and hence can be effective even with limited target data; (iii) improved generalization: The resulting models are intrinsically well-regularized and demonstrate effective generalization even in the challenging partial DA setting, i.e., target domain contains only a subset of the classes observed in the source domain.; and (iv) incremental training: Our approach allows progressive adaptation of models to novel domains (unseen during training) without requiring retraining of the model from scratch.

Abstract The rapid adoption of artificial intelligence methods in healthcare is coupled with the critical need for techniques to rigorously introspect models and thereby ensure that they behave reliably. This has led to the design of explainable AI techniques that uncover the relationships between discernible data signatures and model predictions. In this context, counterfactual explanations that synthesize small, interpretable changes to a given query while producing desired changes in model predictions have become popular. This under-constrained, inverse problem is vulnerable to introducing irrelevant feature manipulations, particularly when the model’s predictions are not well-calibrated. Hence, in this paper, we propose the TraCE (training calibration-based explainers) technique, which utilizes a novel uncertainty-based interval calibration strategy for reliably synthesizing counterfactuals. Given the wide-spread adoption of machine-learned solutions in radiology, our study focuses on deep models used for identifying anomalies in chest X-ray images. Using rigorous empirical studies, we demonstrate the superiority of TraCE explanations over several state-of-the-art baseline approaches, in terms of several widely adopted evaluation metrics. Our findings show that TraCE can be used to obtain a holistic understanding of deep models by enabling progressive exploration of decision boundaries, to detect shortcuts, and to infer relationships between patient attributes and disease severity.

The success of machine learning for scientific discovery normally depends on how well the inherent assumptions match the problem in hand. Here, Thiagarajan et al. alleviate this constraint by allowing the change of optimization criterion in a data-driven approach to emulate complex scientific processes.

Dataset shift refers to the problem where the input data distribution may change over time (e.g., between training and test stages). Since this can be a critical bottleneck in several safety-critical applications such as healthcare, drug-discovery, etc., dataset shift detection has become an important research issue in machine learning. Though several existing efforts have focused on image/video data, applications with graph-structured data have not received sufficient attention. Therefore, in this paper, we investigate the problem of detecting shifts in graph structured data through the lens of statistical hypothesis testing. Specifically, we propose a practical two-sample test based approach for shift detection in large-scale graph structured data. Our approach is very flexible in that it is suitable for both undirected and directed graphs, and eliminates the need for equal sample sizes. Using empirical studies, we demonstrate the effectiveness of the proposed test in detecting dataset shifts. We also corroborate these findings using real-world datasets, characterized by directed graphs and a large number of nodes.

Interpretability has emerged as a crucial aspect of building trust in machine learning systems, aimed at providing insights into the working of complex neural networks that are otherwise opaque to a user. There are a plethora of existing solutions addressing various aspects of interpretability ranging from identifying prototypical samples in a dataset to explaining image predictions or explaining mis-classifications. While all of these diverse techniques address seemingly different aspects of interpretability, we hypothesize that a large family of interepretability tasks are variants of the same central problem which is identifying relative change in a model’s prediction. This paper introduces MARGIN, a simple yet general approach to address a large set of interpretability tasks MARGIN exploits ideas rooted in graph signal analysis to determine influential nodes in a graph, which are defined as those nodes that maximally describe a function defined on the graph. By carefully defining task-specific graphs and functions, we demonstrate that MARGIN outperforms existing approaches in a number of disparate interpretability challenges.