Search for: All records

This paper describes how a system that performs Live Learning can be modified to use radar rather than visible light. The pipelined and iterative machine learning (ML) workflow of this system can operate at low network bandwidths for selective transmission of rare unlabeled events embedded in high-bandwidth real-time sensor data. While radar offers greater range and can overcome the severe signal attenuation experienced by visible light under conditions such as rain or fog, it poses a number of challenges for ML. This paper describes how these challenges can be overcome for Live Learning, and identifies some future directions for research. 

We introduce survival-critical machine learning (SCML), in which a robot encounters dynamically evolving threats that it recognizes via machine learning (ML), and then neutralizes. We model survivability in SCML, and show the value of the recently developed approach of Live Learning. This edge-based ML technique embodies an iterative human-in-the-loop workflow that concurrently enlarges the training set, trains the next model in a sequence of “best-so-far” models, and performs inferencing for both threat detection and pseudo-labeling. We present experimental results using datasets from the domains of drone surveillance, planetary exploration, and underwater sensing to quantify the effectiveness of Live Learning as a mechanism for SCML. 

We present a benchmark-driven experimental study of autonomous drone agility relative to edge offload pipeline attributes. This pipeline includes a monocular gimbal-actuated on-drone camera, hardware RTSP video encoding, 4G LTE wireless network transmission, and computer vision processing on a ground-based GPU-equipped cloudlet. Our parameterized and reproducible agility benchmarks stress the OODA (“Observe, Orient, Decide, Act”) loop of the drone on obstacle avoidance and object tracking tasks. We characterize the latency and throughput of components of this OODA loop through software profiling, and identify opportunities for optimization. 

Drawing on parallels with biological immunity, this paper introduces a new use case for learning at the edge called survival-critical machine learning (SCML). Unlike federated learning, which assumes supervised learning with pre-labeled data, SCML involves semi-supervised learning in streaming settings where labels may need to be obtained at very low network bandwidth and extreme class imbalance. We show that the recently-developed workflow of Live Learning is a good fit for SCML. Starting from a weak bootstrap model, this workflow seamlessly pipelines semi-supervised learning, active learning, and transfer learning, with asynchronous bandwidth-sensitive data transmission for labeling. As improved models evolve at the edge through periodic re-training, the threat detection ability of the SCML system improves. This, in turn, improves the survivability of the host system. 

Edge computing has much lower elasticity than cloud computing because cloudlets have much smaller physical and electrical footprints than a data center. This hurts the scalability of applications that involve low-latency edge offload. We show how this problem can be addressed by leveraging the growing sophistication and compute capability of recent wearable devices. We investigate four Wearable Cognitive Assistance applications on three wearable devices, and show that the technique of offload shaping can significantly reduce network utilization and cloudlet load without compromising accuracy or performance. Our investigation considers the offload shaping strategies of mapping processes to different computing tiers, gating, and decluttering. We find that all three strategies offer a significant bandwidth savings compared to transmitting full camera images to a cloudlet. Two out of the three devices we test are capable of running all offload shaping strategies within a reasonable latency bound. 

A wearable cognitive assistant (WCA) is a computer-based application that guides a user through a task with input from wearable devices with the aid of computational resources in nearby locations (cloudlets). Psychological science informs development of WCAs and is encountering new issues for research. We discuss three relevant research areas: response time, action segmentation, and task comprehension.