2019 International Conference On Computer Aided Design

The Premier Conference Devoted to Technical Innovations in Electronic Design Automation

November 4-7, 2019The Westin Westminster Westminster, CO

v class="event-details"> MP Associates, Inc.

MONDAY November 05, 10:30am - 12:30pm | Riveria
EVENT TYPE: REGULAR SESSION

SESSION 1C
Flexibility Makes Learning Better
Moderator:
Deming Chen - Univ. of Illinois at Urbana-Champaign
Efficient training and deployment of neural networks are critical. The first paper in this session discusses a novel quantization scheme using powers-of-arbitrary-log-bases, and the second paper presents a processing in-DRAM framework for binary CNNs. The third paper proposes AXNet, a neural network based approximate computing allowing holistic end-to-end training. The last paper in this session proposes a scalable-effort CNN (ConvNet) that allows effort-accuracy scalability for classification of data at multi-level abstraction.

1C.1Efficient Hardware Acceleration of CNNs Using Logarithmic Data Representation With Arbitrary Log-Base
 Speaker: Sebastian Vogel - Robert Bosch GmbH
 Authors: Sebastian Vogel - Robert Bosch GmbH
Mengyu Liang - Tech. Univ. of Munich
Andre Guntoro - Robert Bosch GmbH
Walter Stechele - Tech. Univ. of Munich
Gerd Ascheid - RWTH Aachen Univ.
1C.2NID: Processing Binary Convolutional Neural Network in Commodity DRAM
 Speaker: Jaehyeong Sim - Korea Advanced Institute of Science and Technology
 Authors: Jaehyeong Sim - Korea Advanced Institute of Science and Technology
Hoseok Seol - Korea Advanced Institute of Science and Technology
Lee-Sup Kim - Korea Advanced Institute of Science and Technology
1C.3AXNet: ApproXimate Computing Using an End-to-End Trainable Neural Network
 Speaker: Li Jiang - Shanghai Jiao Tong Univ.
 Authors: Zhenghao Peng - Shanghai Jiao Tong Univ.
Li Jiang - Shanghai Jiao Tong Univ.
Xuyang Chen - Shanghai Jiao Tong Univ.
Chengwen Xu - Shanghai Jiao Tong Univ.
Naifeng Jing - Shanghai Jiao Tong Univ.
Xiaoyao Liang - Shanghai Jiao Tong Univ.
Cewu Lu - Shanghai Jiao Tong Univ.
1C.4Scalable-Effort ConvNets for Multilevel Classification
 Speaker: Valentino Peluso - Politecnico di Torino
 Authors: Valentino Peluso - Politecnico di Torino
Andrea Calimera - Politecnico di Torino