Titlebook: Embedded Deep Learning; Algorithms, Architec Bert Moons,Daniel Bankman,Marian Verhelst Book 2019 Springer Nature Switzerland AG 2019 Deep L

1分開

978-3-030-07577-4Springer Nature Switzerland AG 2019

可觸知

Genome

L. F. Clausdorff,K. -P. Hoffmanndiscusses a first . solution for this problem. In this chapter, the wake-up-based detection scenario is generalized to ., where a hierarchy of increasingly complex classifiers, each designed and trained for a specific sub-task, is used to minimize the overall system’s energy cost. An optimal hierarc

genesis

https://doi.org/10.1007/978-3-662-12453-6discusses hardware aware . solutions for this problem. As an introduction to this topic, this chapter gives an overview of existing work in hardware and neural network co-optimizations. Two own contributions in hardware-algorithm optimization are discussed and compared: network quantization either a

transplantation

semiskilled

B. Shah-Derler,E. Wintermantel,S. -W. Hacy through leveraging the three key CNN-characteristics discussed in Chap. .. (a) Inherent CNN parallelism is exploited through a highly parallelized processor architecture that minimizes internal memory bandwidth. (b) Inherent network sparsity in pruned networks and RELU activated feature maps is l

天然熱噴泉

誘拐

嚴(yán)厲批評

L. F. Clausdorff,K. -P. Hoffmannes in a 100-face recognition example. The chips designed in Chap. . are specifically tuned for usage in a hierarchical setup: networks at reduced precision can be used for simple tasks at a high energy efficiency. The chips designed in Chap. . are good candidates for wake-up stages.

ALIBI