Due to the difficulty of obtaining high-quality data in real-world scenarios, datasets inevitably contain noisy labeled data, leading to inefficient data usage and poor model performance. Thus, noisy label detection is an important research topic. Previous efforts mainly focus on noisy label detection on specific datasets that have been collected. Some works select clean samples based on relations between representations during the training process; some works utilize confidence outputs of a pre-trained model for noisy label detection. However, how to perform efficient and fine-grained noisy label detection on constantly arriving datasets in a data lake with a large amount of inventory data has not been explored. The rapidly growing volume and changing distribution of data make conventional methods either incur large computation overhead due to repeated training or become increasingly ineffective on newly arriving data. To address these challenges, in this work, we propose a novel approach ENLD to perform efficient and accurate noisy label detection on incremental datasets. Our extensive experiments demonstrate that ENLD outperforms the next best method in both efficiency and accuracy, which achieves 3.65 ×-4.97× detection speedup and higher average f1 scores with various noise rate settings.

The resource efficiency of video analytics workloads is critical for large-scale deployments on edge nodes and cloud clusters. Recent advanced systems have benefited from techniques including video compression, frame filtering, and deep model acceleration. However, based on our year-long experience of operating a real-time video analytics system on more than 1000 cameras, we identified a previously overlooked bottleneck of end-to-end concurrency: video decoding. To support concurrent video inference at scale, in this work, we investigate a new task, named video packet gating, which selectively filters packets before running a decoder. We propose a novel multi-view embedding approach for video packets and present PacketGame that has both theoretical performance guarantee and practical system designs. Experiments on both public datasets and a real system show PacketGame saves 52.0--79.3% decoding costs and achieves 2.1--4.8× concurrency compared to original workloads. Comparisons with four state-of-the-art complementary methods show the superiority of PacketGame in end-to-end concurrency.

Federated Learning (FL) enables multiple partici-pants to collaboratively train a model in a privacy-preserving way. The performance of the FL model heavily depends on the quality of participants' local data, which makes measuring the contributions of participants an essential task for various purposes, e.g., participant selection and reward allocation. The Shapley value is widely adopted by previous work for contribution assessment, which, however, requires repeatedly leave-one-out retraining and thus incurs the prohibitive cost for FL. In this paper, we propose a highly efficient approach, named DIG-FL, to estimate the Shapley value of each participant without any model retraining. It's worth noting that our approach is applicable to both vertical federated learning (VFL) and horizontal federated learning (HFL), and we provide concrete design for VFL and HFL. In addition, we propose a DIG-FL based reweight mechanism to improve the model training in terms of accuracy and convergence speed by dynamically adjusting the weights of participants according to their per-epoch contributions, and theoretically analyze the convergence speed. Our extensive evaluations on 14 public datasets show that the estimated Shapley value is very close to the actual Shapley value with Pearson's correlation coefficient up to 0.987, while the cost is orders of magnitude smaller than state-of-the-art methods. When there are more than 80% participants holding low-quality data, by dynamically adjusting the weights, DIG-FL can effectively accelerate the convergence and improve the model accuracy.

The widespread creative application and smart devices call for convenient and secure interaction with human users. We propose, design, and implement a smartwatch-based two-factor real-time identification and authentication system named ThumbUp, where smartwatch users can identify and authenticate themselves by some simple hand and finger gestures, such as thumb-up. ThumbUp leverages the signal collected from the Inertial Measurement Unit (IMU) in Commercial Off-The-Shelf (COTS) smart devices and discovers the unique fingerprint pattern produced by each user’s simple hand gestures using a carefully crafted deep learning model. We implement our system and conduct extensive experiments to evaluate its efficacy and efficiency with 65 different users over a period of more than 3 months. It reaches an accuracy of 97% for identification, and EER 0.014 for authentication using only one simple gesture. We also survey the users’ acceptance of our system and discuss how the proficiency of gestures affects authentication accuracy.

Nowadays, there is a growing demand for sharing multimedia data among participants in the same activity. With existing social applications, users need to conduct friending and data sharing operations manually, which is troublesome due to changing attendees and highly diverse data content of different activities. To tackle this issue, in this work we propose a novel system SHAD to achieve privacy-friendly shared activity detection and multimedia data auto-sharing based on users' historical multimodal data. Facing noisy, incomplete and asynchronous data, as well as inaccurate recognition results of machine learning models, we design an algorithm to aggregate multimodal data relevant to the same activity and propose an activity-semantic graph to comprehensively characterize each activity by fusing knowledge of multimodal data. Based on the activity-semantic graph, the privacy-preserving shared activity detection and data sharing method is designed, which protects both raw data and semantic information of data. We implemented our system and conducted comprehensive evaluations with real-life multimodal data (including photos and motion sensor data). The results show the efficacy of our system. We can achieve 94.9% precision and 91.5% recall for shared activity detection.

Soft compliant materials and novel actuation mechanisms ensure flexible motions and high adaptability for soft robots, but also increase the difficulty and complexity of constructing control systems. In this work, we provide an efficient control algorithm for a multi-segment extensible soft arm in 2D plane. The algorithm separate the inverse kinematics into two levels. The first level employs gradient descent to select optimized arm's pose (from task space to configuration space) according to designed cost functions. With consideration of viscoelasticity, the second level utilizes neural networks to figure out the pressures from each segment's pose (from configuration space to actuation space). In experiments with a physical prototype, the control accuracy and effectiveness are validated, where the control algorithm is further improved by an optional feedback strategy.

Most control methods of soft manipulators are developed based on physical models derived from mathematical analysis or learning methods. However, due to internal nonlinearity and external uncertain disturbances, it is difficult to build an accurate model, further, these methods lack robustness and portability among different prototypes. In this work, we propose a model-free control method based on reinforcement learning and implement it on a multi-segment soft manipulator in 2D plane, which focuses on the learning of control strategy rather than the physical model. The control strategy is validated to be effective and robust in prototype experiments, where we design a simulation method to speed up the training process.