Search for: All records

Selecting representative samples plays an indispensable role in many machine learning and computer vision applications under limited resources (e.g., limited communication bandwidth and computational power). Determinantal Point Process (DPP) is a widely used method for selecting the most diverse representative samples that can summarize a dataset. However, its adaptability to different tasks remains an open challenge, as it is challenging for DPP to perform task-specific tuning. In contrast, Rate-Distortion (RD) theory provides a way to measure task-specific diversity. However, optimizing RD for a data selection problem remains challenging because the quantity that needs to be optimized is the index set of the selected samples. To tackle these challenges, we first draw an inherent relationship between DPP and RD theory. Our theoretical derivation paves the way for taking advantage of both RD and DPP for a task-specific data selection. To this end, we propose a novel method for task-specific data selection for multi-level classification tasks, named RD-DPP. Empirical studies on seven different datasets using five benchmark models demonstrate the effectiveness of the proposed RD-DPP method. Our method also outperforms recent strong competing methods, while exhibiting high generalizability to a variety of learning tasks. 

Determinantal Point Process (DPP) is a powerful technique to enhance data diversity by promoting the repulsion of similar elements in the selected samples. Particularly, DPP-based Maximum A Posteriori (MAP) inference is used to identify subsets with the highest diversity. However, a commonly adopted presumption of all data samples being available at one point hinders its applicability to real-world scenarios where data samples are distributed across distinct sources with intermittent and bandwidth-limited connections. This paper proposes a distributed version of DPP inference to enhance multi-source data diversification under limited communication budgets. First, we convert the lower bound of the diversity-maximized distributed sample selection from matrix determinant optimization to a simpler form of the sum of individual terms. Next, a determinant-preserved sparse representation of selected samples is formed by the sink as a surrogate for collected samples and sent back to sources as lightweight messages to eliminate the need for raw data exchange. Our approach is inspired by the channel orthogonalization process of Multiple-Input Multiple-Output (MIMO) systems based on the Channel State Information (CSI). Extensive experiments verify the superiority of our scalable method over the most commonly used data selection methods, including GreeDi, Greedymax, random selection, and stratified sampling by a substantial gain of at least 12% reduction in Relative Diversity Error (RDE). This enhanced diversity translates to a substantial improvement in the performance of various downstream learning tasks, including multi-level classification (2%-4% gain in accuracy), object detection (2% gain in mAP), and multiple-instance learning (1.3% gain in AUC). 

Deep neural networks, including transformers and convolutional neural networks (CNNs), have significantly improved multivariate time series classification (MTSC). However, these methods often rely on supervised learning, which does not fully account for the sparsity and locality of patterns in time series data (e.g., quantification of diseases-related anomalous points in ECG and abnormal detection in signal). To address this challenge, we formally discuss and reformulate MTSC as a weakly supervised problem, introducing a novel multiple-instance learning (MIL) framework for better localization of patterns of interest and modeling time dependencies within time series. Our novel approach, TimeMIL, formulates the temporal correlation and ordering within a time-aware MIL pooling, leveraging a tokenized transformer with a specialized learnable wavelet positional token. The proposed method surpassed 26 recent state-of-the-art MTSC methods, underscoring the effectiveness of the weakly supervised TimeMIL in MTSC. The code is available https://github.com/xiwenc1/TimeMIL. 

Many AI platforms, including traffic monitoring systems, use Federated Learning (FL) for decentralized sensor data processing for learning-based applications while preserving privacy and ensuring secured information transfer. On the other hand, applying supervised learning to large data samples, like high-resolution images requires intensive human labor to label different parts of a data sample. Multiple Instance Learning (MIL) alleviates this challenge by operating over labels assigned to the ’bag’ of instances. In this paper, we introduce Federated Multiple-Instance Learning (FedMIL). This framework applies federated learning to boost the training performance in video-based MIL tasks such as vehicle accident detection using distributed CCTV networks. However, data sources in decentralized settings are not typically Independently and Identically Distributed (IID), making client selection imperative to collectively represent the entire dataset with minimal clients. To address this challenge, we propose DPPQ, a framework based on the Determinantal Point Process (DPP) with a quality-based kernel to select clients with the most diverse datasets that achieve better performance compared to both random selection and current DPP-based client selection methods even with less data utilization in the majority of non-IID cases. This offers a significant advantage for deployment on edge devices with limited computational resources, providing a reliable solution for training AI models in massive smart sensor networks. 

Current forest monitoring technologies including satellite remote sensing, manned/piloted aircraft, and observation towers leave uncertainties about a wildfire’s extent, behavior, and conditions in the fire’s near environment, particularly during its early growth. Rapid mapping and real-time fire monitoring can inform in-time intervention or management solutions to maximize beneficial fire outcomes. Drone systems’ unique features of 3D mobility, low flight altitude, and fast and easy deployment make them a valuable tool for early detection and assessment of wildland fires, especially in remote forests that are not easily accessible by ground vehicles. In addition, the lack of abundant, well-annotated aerial datasets – in part due to unmanned aerial vehicles’ (UAVs’) flight restrictions during prescribed burns and wildfires – has limited research advances in reliable data-driven fire detection and modeling techniques. While existing wildland fire datasets often include either color or thermal fire images, here we present (1) a multi-modal UAV-collected dataset of dual-feed side-by-side videos including both RGB and thermal images of a prescribed fire in an open canopy pine forest in Northern Arizona and (2) a deep learning-based methodology for detecting fire and smoke pixels at accuracy much higher than the usual single-channel video feeds. The collected images are labeled to “fire” or “no-fire” frames by two human experts using side-by-side RGB and thermal images to determine the label. To provide context to the main dataset’s aerial imagery, the included supplementary dataset provides a georeferenced pre-burn point cloud, an RGB orthomosaic, weather information, a burn plan, and other burn information. By using and expanding on this guide dataset, research can develop new data-driven fire detection, fire segmentation, and fire modeling techniques.