Evolutionary Optimization of Deep Learning Activation Functions

被引:22
作者
Bingham, Garrett [1 ,2 ]
Macke, William [1 ]
Miikkulainen, Risto [1 ,2 ]
机构
[1] Univ Texas Austin, Austin, TX 78712 USA
[2] Cognizant Technol Solut, San Francisco, CA 94111 USA
来源
GECCO'20: PROCEEDINGS OF THE 2020 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE | 2020年
关键词
Activation functions; evolutionary algorithms; metalearning;
D O I
10.1145/3377930.3389841
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The choice of activation function can have a large effect on the performance of a neural network. While there have been some attempts to hand-engineer novel activation functions, the Rectified Linear Unit (ReLU) remains the most commonly-used in practice. This paper shows that evolutionary algorithms can discover novel activation functions that outperform ReLU. A tree-based search space of candidate activation functions is defined and explored with mutation, crossover, and exhaustive search. Experiments on training wide residual networks on the CIFAR-10 and CIFAR-100 image datasets show that this approach is effective. Replacing ReLU with evolved activation functions results in statistically significant increases in network accuracy. Optimal performance is achieved when evolution is allowed to customize activation functions to a particular task; however, these novel activation functions are shown to generalize, achieving high performance across tasks. Evolutionary optimization of activation functions is therefore a promising new dimension of metalearning in neural networks.
引用
收藏
页码:289 / 296
页数:8
相关论文
共 23 条
[1]  
Abadi M, 2016, PROCEEDINGS OF OSDI'16: 12TH USENIX SYMPOSIUM ON OPERATING SYSTEMS DESIGN AND IMPLEMENTATION, P265
[2]  
[Anonymous], 2017, P INT C LEARN REPR T
[3]  
[Anonymous], 2018, arXiv
[4]  
[Anonymous], 2013, PROC 30 INT C MACH L
[5]  
[Anonymous], 2016, P BRIT MACH VIS C
[6]  
[Anonymous], 2010, P 27 INT C MACH LEAR, DOI 10.5555/3104322.3104425
[7]  
[Anonymous], ARXIV
[8]   Progressive Differentiable Architecture Search: Bridging the Depth Gap between Search and Evaluation [J].
Chen, Xin ;
Xie, Lingxi ;
Wu, Jun ;
Tian, Qi .
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019), 2019, :1294-1303
[9]  
Elsken T, 2019, J MACH LEARN RES, V20
[10]  
Feurer M., 2015, 29 AAAI C ART INT