A Comprehensive Survey on Training Acceleration for Large Machine Learning Models in IoT

Wang, Haozhao; Qu, Zhihao; Zhou, Qihua; Zhang, Haobo; Luo, Boyuan; Xu, Wenchao; Guo, Song; Li, Ruixuan

doi:10.1109/jiot.2021.3111624

Cited by 25 publications

(11 citation statements)

References 113 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From Section 2, it follows that any candidate controller that satisfies the adjoint equations and the transversality conditions, namely, (10) and (11), produces an algorithm primitive (cf. ( 5)).…”

Section: A Deductive Approach To Designing Algorithm Primitives and A...mentioning

confidence: 99%

“…The widespread and practical applications of ML to such problems as image recognition, cancer detection, self-driving automobiles and other big-data problems [9] changed prevailing attitudes about previously-discarded optimization algorithms [6,7,8]. CD algorithms are now more widely used than ever before in such diverse application as medical imaging [10], internet-of-things [11], computational biology [12], and much more [6,7]. By absorbing it under the emerging unified theory for optimization, we anticipate new formalisms for those ML algorithms that utilize CD methods.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Derivation of Coordinate Descent Algorithms from Optimal Control Theory

Ross

2023

Oper. Res. Forum

View full text Add to dashboard Cite

Recently, it was posited that disparate optimization algorithms may be coalesced in terms of a central source emanating from optimal control theory. Here we further this proposition by showing how coordinate descent algorithms may be derived from this emerging new principle. In particular, we show that basic coordinate descent algorithms can be derived using a maximum principle and a collection of max functions as "control" Lyapunov functions. The convergence of the resulting coordinate descent algorithms is thus connected to the controlled dissipation of their corresponding Lyapunov functions. The operational metric for the search vector in all cases is given by the Hessian of the convex objective function.

show abstract

Section: A Deductive Approach To Designing Algorithm Primitives and A...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Derivation of Coordinate Descent Algorithms from Optimal Control Theory

Ross

2023

Oper. Res. Forum

View full text Add to dashboard Cite

show abstract

“…Consequently, edge computing capabilities are being boosted and explored to deliver better intelligence engine inference services to end-users [33]. For instance, in [34], they worked on accelerating the training process of large machine learning models in IoT to meet the hardware limitations.…”

Section: B Internet Of Bodiesmentioning

confidence: 99%

Bindi: Affective Internet of Things to Combat Gender-Based Violence

Miranda

Rituerto-González

Luis-Mingueza

et al. 2022

IEEE Internet Things J.

View full text Add to dashboard Cite

The main research motivation of this paper is the fight against gender-based violence and achieving gender equality from a technological perspective. The solution proposed in this work goes beyond currently existing panic buttons, needing to be manually operated by the victims under difficult circumstances. Instead, Bindi, our end-to-end autonomous multimodal system, relies on artificial intelligence methods to automatically identify violent situations, based on detecting fear-related emotions, and trigger a protection protocol, if necessary. To this end, Bindi integrates modern state-of-the-art technologies, such as the Internet of Bodies, affective computing, and cyber-physical systems, leveraging i) affective IoT with auditory and physiological commercial off-the-shelf smart sensors embedded in wearable devices, ii) hierarchical multisensorial information fusion, and iii) the edge-fog-cloud IoT architecture. This solution is evaluated using our own dataset named WEMAC, a very recently collected and freely-available collection of data comprising the auditory and physiological responses of 47 women to several emotions elicited by using a virtual reality environment. On this basis, this work provides an analysis of multimodal late fusion strategies to combine the physiological and speech data processing pipelines to identify the best intelligence engine strategy for Bindi. In particular, the best data fusion strategy reports an overall fear classification accuracy of 63.61% for a subject-independent approach. Both a power consumption study and an audio data processing pipeline to detect violent acoustic events complement this analysis. This research is intended as an initial multimodal baseline that facilitates further work with real-life elicited fear in women.

show abstract

“…Each gadget may have its own utility that benefits the overall network. This is the foundation of IoT technology [33], [68], [100], [112], [127], [152], [280].…”

Section: Future Directionsmentioning

confidence: 99%

Insights on the internet of things: past, present, and future directions

Sutikno

Thalmann

2022

TELKOMNIKA

View full text Add to dashboard Cite

The internet of things (IoT) is rapidly expanding and improving operations in a wide range of real-world applications, from consumer IoT and enterprise IoT to manufacturing and industrial IoT (IIoT). Consumer markets, wearable devices, healthcare, smart buildings, agriculture, and smart cities are just a few examples. This paper discusses the current state of the IoT ecosystem, its primary applications and benefits, important architectural stages, some of the problems and challenges it faces, and its future. This paper explains how an appropriate IoT architecture that saves data, analyzes it, and recommends corrective action improves the process's ground reality. The IoT system architecture is divided into three layers: device, gateway, and platform. This then cascades into the four stages of the IoT architectural layout: sensors and actuators; gateways and data acquisition systems; edge IT data processing; and datacenter and cloud, which use high-end apps to collect data, evaluate it, process it, and provide remedial solutions. This elegant combination provides excellent value in automatic action. In the future, IoT will continue to serve as the foundation for many technologies. Machine learning will become more popular in the coming years as IoT networks take center stage in a variety of industries.

show abstract

A Comprehensive Survey on Training Acceleration for Large Machine Learning Models in IoT

Cited by 25 publications

References 113 publications

Derivation of Coordinate Descent Algorithms from Optimal Control Theory

Derivation of Coordinate Descent Algorithms from Optimal Control Theory

Bindi: Affective Internet of Things to Combat Gender-Based Violence

Insights on the internet of things: past, present, and future directions

Contact Info

Product

Resources

About