Search for: All records

Recognizing food types through sensor signals for unseen users remains remarkably challenging, despite extensive recent studies. The efficacy of prior machine learning techniques is dwarfed by giant variations of data collected from multiple participants, partly because users have varied chewing habits and wear sensor devices in various manners. This work treats the problem as an instance of the domain adaptation problem, where each user represents a domain. We develop the first multi-source domain adaptation (MSDA) method for food-typing recognition, which consists of three major components: stratified normalization, a multi-source domain adaptor, and adaptive ensemble learning. New techniques are developed for each component. Using a real-world dataset comprised of 15 participants, we demonstrate that our method achieves\(1.33\times\)to\(2.13\times\)improvement in accuracy compared with nine state-of-the-art MSDA baselines. Additionally, we perform an in-depth ablation study to examine the behavior of each component and confirm their efficacy.

 

Though many compilation and runtime systems have been developed for DNNs in recent years, the focus has largely been on static DNNs. Dynamic DNNs, where tensor shapes and sizes and even the set of operators used are dependent upon the input and/or execution are becoming common. This paper presents SoD2, a comprehensive framework for optimizing Dynamic DNNs. The basis of our approach is a classification of common operators that form DNNs, and the use of this classification towards a Rank and Dimension Propagation (RDP) method. This framework statically determines the shapes of operators as known constants, symbolic constants, or operations on these. Next, using RDP we enable a series of optimizations, like fused code generation, execution (order) planning, and even runtime memory allocation plan generation. By evaluating the framework on 10 emerging Dynamic DNNs and comparing it against several existing systems, we demonstrate both reductions in execution latency and memory requirements, with RDP-enabled key optimizations responsible for much of the gains. 

Data redundancy is ubiquitous in the inputs and intermediate results of Deep Neural Networks (DNN) . It offers many significant opportunities for improving DNN performance and efficiency and has been explored in a large body of work. These studies have scattered in many venues across several years. The targets they focus on range from images to videos and texts, and the techniques they use to detect and exploit data redundancy also vary in many aspects. There is not yet a systematic examination and summary of the many efforts, making it difficult for researchers to get a comprehensive view of the prior work, the state of the art, differences and shared principles, and the areas and directions yet to explore. This article tries to fill the void. It surveys hundreds of recent papers on the topic, introduces a novel taxonomy to put the various techniques into a single categorization framework, offers a comprehensive description of the main methods used for exploiting data redundancy in improving multiple kinds of DNNs on data, and points out a set of research opportunities for future exploration. 

2D van der Waals (vdW) magnets open landmark horizons in the development of innovative spintronic device architectures. However, their fabrication with large scale poses challenges due to high synthesis temperatures (>500 °C) and difficulties in integrating them with standard complementary metal‐oxide semiconductor (CMOS) technology on amorphous substrates such as silicon oxide (SiO₂) and silicon nitride (SiN_x). Here, a seeded growth technique for crystallizing CrTe₂films on amorphous SiN_x/Si and SiO₂/Si substrates with a low thermal budget is presented. This fabrication process optimizes large‐scale, granular atomic layers on amorphous substrates, yielding a substantial coercivity of 11.5 kilo‐oersted, attributed to weak intergranular exchange coupling. Field‐driven Néel‐type stripe domain dynamics explain the amplified coercivity. Moreover, the granular CrTe₂devices on Si wafers display significantly enhanced magnetoresistance, more than doubling that of single‐crystalline counterparts. Current‐assisted magnetization switching, enabled by a substantial spin–orbit torque with a large spin Hall angle (85) and spin Hall conductivity (1.02 ×  10⁷ℏ/2e  Ω⁻¹  m⁻¹), is also demonstrated. These observations underscore the proficiency in manipulating crystallinity within integrated 2D magnetic films on Si wafers, paving the way for large‐scale batch manufacturing of practical magnetoelectronic and spintronic devices, heralding a new era of technological innovation.