Enhanced methods for Evolution in-Materio Processors

被引:1
作者
Jones, Benedict A. H. [1 ]
Al Moubayed, Noura [2 ]
Zeze, Dagou A. [1 ]
Groves, Chris [1 ]
机构
[1] Univ Durham, Dept Engn, Durham DH1 3LE, England
[2] Univ Durham, Dept Comp Sci, Durham DH1 3LE, England
来源
2021 INTERNATIONAL CONFERENCE ON REBOOTING COMPUTING (ICRC 2021) | 2021年
关键词
Batching; binary cross entropy; evolution in-materio processors; evolutionary materials; evolvable processors; material kernel; DIFFERENTIAL EVOLUTION; NETWORK;
D O I
10.1109/ICRC53822.2021.00026
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Evolution-in-Materio (EiM) is an unconventional computing paradigm, which uses an Evolutionary Algorithm (EA) to configure a material's parameters so that it can perform a computational task. While EiM processors show promise, slow manufacturing and physical experimentation hinder their development. Simulations based on a physical model were used to efficiently investigate three specific enhancements to EiM processors which operate as classifiers. Firstly, an adapted Differential Evolution algorithm that includes batching and a validation dataset. This allows more generational updates and a validation metric which could tune hyper-parameters. Secondly, the introduction of Binary Cross Entropy as an objective function for the EA, a continuous fitness metric with several advantages over the commonly used classification error objective function. Finally, the use of regression to quickly assess the material processor's output states and produce an optimal readout layer, a significant improvement over fixed or evolved interpretation schemes which can 'hide' the true performance of a material processor. Together these enhancements provide guidance on the production of more flexible, better performing, and robust EiM processors.
引用
收藏
页码:109 / 118
页数:10
相关论文
共 50 条
[21]   An enhanced utilization mechanism of population information for Differential evolution [J].
Chi Shao ;
Yiqiao Cai ;
Shunkai Fu ;
Jing Li ;
Wei Luo .
Evolutionary Intelligence, 2022, 15 :2247-2259
[22]   An enhanced utilization mechanism of population information for Differential evolution [J].
Shao, Chi ;
Cai, Yiqiao ;
Fu, Shunkai ;
Li, Jing ;
Luo, Wei .
EVOLUTIONARY INTELLIGENCE, 2022, 15 (04) :2247-2259
[23]   Optimizing Polynomial Window Functions by Enhanced Differential Evolution [J].
Jia, Dongli ;
Zheng, Guoxin ;
Zhu, Yazhou ;
Zhang, Li .
PROCEEDINGS OF INTERNATIONAL SYMPOSIUM ON COMPUTER SCIENCE AND COMPUTATIONAL TECHNOLOGY (ISCSCT 2009), 2009, :218-221
[24]   CUDA BASED ENHANCED DIFFERENTIAL EVOLUTION: A COMPUTATIONAL ANALYSIS [J].
Davendra, Donald ;
Gaura, Jan ;
Bialic-Davendra, Magdalena ;
Senkerik, Roman .
PROCEEDINGS 26TH EUROPEAN CONFERENCE ON MODELLING AND SIMULATION ECMS 2012, 2012, :399-+
[25]   Enhanced artificial bee colony algorithm through differential evolution [J].
Gao, Wei-feng ;
Huang, Ling-ling ;
Wang, Jue ;
Liu, San-yang ;
Qin, Chuan-dong .
APPLIED SOFT COMPUTING, 2016, 48 :137-150
[26]   Flow shop scheduling using enhanced differential evolution algorithm [J].
Davendra, Donald ;
Onwubolu, Godfrey .
21ST EUROPEAN CONFERENCE ON MODELLING AND SIMULATION ECMS 2007: SIMULATIONS IN UNITED EUROPE, 2007, :259-+
[27]   Enhanced versions of differential evolution: state-of-the-art survey [J].
Mashwani, Wali Khan .
INTERNATIONAL JOURNAL OF COMPUTING SCIENCE AND MATHEMATICS, 2014, 5 (02) :107-126
[28]   Enhanced Differential Evolution by Dynamic Selection Framework of Mutation Operator [J].
Xia Dahai ;
Lin Song ;
Gu Wei ;
Xiong Caiquan .
2018 8TH INTERNATIONAL CONFERENCE ON ELECTRONICS INFORMATION AND EMERGENCY COMMUNICATION (ICEIEC), 2018, :243-246
[29]   Enhanced Differential Evolution Entirely Parallel Method for Biomedical Applications [J].
Kozlov, Konstantin ;
Ivanisenko, Nikita ;
Ivanisenko, Vladimir ;
Kolchanov, Nikolay ;
Samsonova, Maria ;
Samsonov, Alexander M. .
PARALLEL COMPUTING TECHNOLOGIES (PACT 2013), 2013, 7979 :409-416
[30]   On the Use of Repair Methods in Differential Evolution for Dynamic Constrained Optimization [J].
Ameca-Alducin, Maria-Yaneli ;
Hasani-Shoreh, Maryam ;
Neumann, Frank .
APPLICATIONS OF EVOLUTIONARY COMPUTATION, EVOAPPLICATIONS 2018, 2018, 10784 :832-847