A5102778981- Parallel Computing and Optimization Techniques
- Language, Metaphor, and Cognition
- Blood Pressure and Hypertension Studies
- Diabetes, Cardiovascular Risks, and Lipoproteins
- Robotics and Sensor-Based Localization
- Robotics and Automated Systems
- Advanced Memory and Neural Computing
- Advanced Data Storage Technologies
- Advanced Neural Network Applications
- Robotic Path Planning Algorithms
- Ferroelectric and Negative Capacitance Devices
- Lipoproteins and Cardiovascular Health
Yonsei University
2020-2021
This paper presents a hardware management technique that enables energy-efficient acceleration of deep neural networks (DNNs) on realtime-constrained embedded edge devices. It becomes increasingly common for devices to incorporate dedicated accelerators processing. The execution in general follows host-device model, where CPUs offload computations (e.g., matrix and vector calculations) the datapath-optimized executions. Such serialized is simple implement manage, but it wasteful...
This article presents a benchmark suite named <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Nebula</i> that implements lightweight neural network benchmarks. Recent networks tend to form deeper and sizable enhance accuracy applicability. However, the massive volume of heavy makes them highly challenging use in conventional research environments such as microarchitecture simulators. We notice computations are mainly comprised matrix vector...
Visual arguments, often used in advertising or social causes, rely on images to persuade viewers do believe something. Understanding these arguments requires selective vision: only specific visual stimuli within an image are relevant the argument, and relevance can be understood context of a broader argumentative structure. While readily appreciated by human audiences, we ask: today's AI capable similar understanding? We collect release VisArgs, annotated corpus designed make explicit...
Real-life robot navigation involves more than just reaching a destination; it requires optimizing movements while addressing scenario-specific goals. An intuitive way for humans to express these goals is through abstract cues like verbal commands or rough sketches. Such human guidance may lack details be noisy. Nonetheless, we expect robots navigate as intended. For interpret and execute instructions in line with expectations, they must share common understanding of basic concepts humans. To...