On parallel online learning for adaptive embedded systems
2014 (English)In: Advancing Embedded Systems and Real-Time Communications with Emerging Technologies, IGI Global, 2014, 262-280 p.Chapter in book (Other academic)Text
This chapter considers parallel implementation of the online multi-label regularized least-squares machinelearning algorithm for embedded hardware platforms. The authors focus on the following properties required in real-time adaptive systems: learning in online fashion, that is, the model improves with new data but does not require storing it; the method can fully utilize the computational abilities of modern embedded multi-core computer architectures; and the system efficiently learns to predict several labels simultaneously. They demonstrate on a hand-written digit recognition task that the online algorithm converges faster, with respect to the amount of training data processed, to an accurate solution than a stochastic gradient descent based baseline. Further, the authors show that our parallelization of the method scales well on a quad-core platform. Moreover, since Network-on-Chip (NoC) has been proposed as a promising candidate for future multi-core architectures, they implement a NoC system consisting of 16 cores. The proposed machine learning algorithm is evaluated in the NoC platform. Experimental results show that, by optimizing the cache behaviour of the program, cache/memory efficiency can improve significantly. Results from the chapter provide a guideline for designing future embedded multicore machine learning devices.
Place, publisher, year, edition, pages
IGI Global, 2014. 262-280 p.
IdentifiersURN: urn:nbn:se:kth:diva-181269DOI: 10.4018/978-1-4666-6034-2.ch011ScopusID: 2-s2.0-84945990377ISBN: 9781466660366OAI: oai:DiVA.org:kth-181269DiVA: diva2:902626
QC 201602112016-02-112016-01-292016-02-11Bibliographically approved