More Like this

1. Generative-Discriminative Complementary Learning

   https://doi.org/10.1609/aaai.v34i04.6126  

   Xu, Yanwu; Gong, Mingming; Chen, Junxiang; Liu, Tongliang; Zhang, Kun; Batmanghelich, Kayhan (June 2020, Proceedings of the AAAI Conference on Artificial Intelligence)

   The majority of state-of-the-art deep learning methods are discriminative approaches, which model the conditional distribution of labels given inputs features. The success of such approaches heavily depends on high-quality labeled instances, which are not easy to obtain, especially as the number of candidate classes increases. In this paper, we study the complementary learning problem. Unlike ordinary labels, complementary labels are easy to obtain because an annotator only needs to provide a yes/no answer to a randomly chosen candidate class for each instance. We propose a generative-discriminative complementary learning method that estimates the ordinary labels by modeling both the conditional (discriminative) and instance (generative) distributions. Our method, we call Complementary Conditional GAN (CCGAN), improves the accuracy of predicting ordinary labels and is able to generate high-quality instances in spite of weak supervision. In addition to the extensive empirical studies, we also theoretically show that our model can retrieve the true conditional distribution from the complementarily-labeled data. 

   more » « less
2. Active Deep Learning for Activity Recognition with Context Aware Annotator Selection

   https://doi.org/10.1145/3292500.3330688  

   Hossain, H M; Roy, Nirmalya (January 2019, Proceedings of the 25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Anchorage, Alaska, August 2019)

   Machine learning models are bounded by the credibility of ground truth data used for both training and testing. Regardless of the problem domain, this ground truth annotation is objectively manual and tedious as it needs considerable amount of human intervention. With the advent of Active Learning with multiple annotators, the burden can be somewhat mitigated by actively acquiring labels of most informative data instances. However, multiple annotators with varying degrees of expertise poses new set of challenges in terms of quality of the label received and availability of the annotator. Due to limited amount of ground truth information addressing the variabilities of Activity of Daily Living (ADLs), activity recognition models using wearable and mobile devices are still not robust enough for real-world deployment. In this paper, we propose an active learning combined deep model which updates its network parameters based on the optimization of a joint loss function. We then propose a novel annotator selection model by exploiting the relationships among the users while considering their heterogeneity with respect to their expertise, physical and spatial context. Our proposed model leverages model-free deep reinforcement learning in a partially observable environment setting to capture the actionreward interaction among multiple annotators. Our experiments in real-world settings exhibit that our active deep model converges to optimal accuracy with fewer labeled instances and achieves 8% improvement in accuracy in fewer iterations. 

   more » « less
3. Towards Joint Segmentation and Active Learning for Block-Structured Data Streams

   Conrad Holtsclaw, Meet P. (August 2019, The ACM SIGKDD Conference on Knowledge Discovery and Data Mining Workshop on Data Collection, Curation, and Labeling for Mining and Learning)

   Stream-based active learning methods assume that data instances arrive in sequence and the decision must be made to query an instance or not as it arrives. In mobile health and human activity recognition, the data stream is often block-structured where instances in the same block have the same label, but the boundaries between blocks are unobserved. In this paper, we propose an approach to active learning in this setting where we simultaneously learn to segment the stream while learning an instance-level discriminative classifier. We show that by propagating collected labels into inferred segments, we can learn improved predictive models with significantly fewer queries. 

   more » « less
4. Action Completeness Modeling with Background Aware Networks for Weakly-Supervised Temporal Action Localization

   Moniruzzaman, M; Yin, Z; He, Z; Qin, R; Leu, M (October 2020, Proceedings of ACM Multimedia Conference 2020)

   null (Ed.)

   The state-of-the-art of fully-supervised methods for temporal action localization from untrimmed videos has achieved impressive results. Yet, it remains unsatisfactory for the weakly-supervised temporal action localization, where only video-level action labels are given without the timestamp annotation on when the actions occur. The main reason comes from that, the weakly-supervised networks only focus on the highly discriminative frames, but there are some ambiguous frames in both background and action classes. The ambiguous frames in background class are very similar to the real actions, which may be treated as target actions and result in false positives. On the other hand, the ambiguous frames in action class which possibly contain action instances, are prone to be false negatives by the weakly-supervised networks and result in a coarse localization. To solve these problems, we introduce a novel weakly-supervised Action Completeness Modeling with Back- ground Aware Networks (ACM-BANets). Our Background Aware Network (BANet) contains a weight-sharing two-branch architecture, with an action guided Background aware Temporal Attention Module (B-TAM) and an asymmetrical training strategy, to suppress both highly discriminative and ambiguous background frames to remove the false positives. Our action completeness modeling contains multiple BANets, and the BANets are forced to discover different but complementary action instances to completely localize the action instances in both highly discriminative and ambiguous action frames. In the 𝑖-th iteration, the 𝑖-th BANet discovers the discriminative features, which are then erased from the feature map. The partially-erased feature map is fed into the (i+1)-th BANet of the next iteration to force this BANet to discover discriminative features different from the 𝑖-th BANet. Evaluated on two challenging untrimmed video datasets, THUMOS14 and ActivityNet1.3, our approach outperforms all the current weakly-supervised methods for temporal action localization. 

   more » « less
5. A Multimodal Deep Learning Approach for Automated Detection and Characterization of Distinctly Salient Features of Alzheimers Disease

   https://doi.org/10.1109/ISBI53787.2023.10230525  

   Batta, Ishaan; Abrol, Anees; Calhoun, Vince (April 2023, IEEE)

   Neurological disorders generally involve multiple kinds of changes in the functional and structural properties of the brain. In this study, we develop a CNN-based multimodal deep learning pipeline by exploiting both functional and structural neuroimaging features to generate full-brain maps that encode significant differences between patient groups and between modalities in terms of their distinctive contribution towards diagnostic classification of Alzheimer’s disease. Through a repeated cross-validation procedure and robust statistical analysis, we show that our approach can be used to encode highly discriminative and abstract information from full-brain data, while also retaining the ability to identify and categorize significantly contributing voxel-level features based on their salient strength in various diagnostic and modality-related contexts. Our results on an Alzheimer’s disease classification task show that such approaches can be used for creating more elaborately defined biomarkers for brain disorders. 

   more » « less