Reliable and Fast Recurrent Neural Network Architecture Optimization

This article introduces Random Error Sampling-based Neuroevolution (RESN), a novel automatic method to optimize recurrent neural network architectures. RESN combines an evolutionary algorithm with a training-free evaluation approach. The results show that RESN achieves state-of-the-art error performance while reducing by half the computational time.

cs.NE cs.LG Neural Networks and Applications Machine Learning and ELM Energy Load and Power Forecasting Computer Science Engineering

View graph Source

Paper

References (14)

01Random error sampling-based recurrent neural network architecture optimizationAndrés Camero, J. Toutouh, E. Alba2019 · Engineering applications of artificial intelligence · 25 citations

02Evolving Recurrent Neural Networks for Time Series Data Prediction of Coal Plant ParametersA. Elsaid, Steven A. Benson, Shuchita Patwardhan et al.2019 · EvoApplications · 30 citations

03Investigating recurrent neural network memory structures using neuro-evolutionAlexander G. Ororbia, A. Elsaid, Travis Desell2019 · Annual Conference on Genetic and Evolutionary Computation · 62 citations In Library

04Waste Generation Prediction in Smart Cities Through Deep NeuroevolutionAndrés Camero, J. Toutouh, Javier Ferrer et al.2018 · Ibero-American Congress of Smart Cities · 12 citations

05BIN-CT: Urban Waste Collection based in Predicting the Container Fill LevelJavier Ferrer, E. Alba2018 · Biosyst. · 44 citations In Library

06Low-Cost Recurrent Neural Network Expected Performance EvaluationAndrés Camero, J. Toutouh, E. Alba2018 · arXiv.org · 20 citations In Library

07Evolutionary architecture search for deep multitask networksJ. Liang, Elliot Meyerson, R. Miikkulainen2018 · Annual Conference on Genetic and Evolutionary Computation · 128 citations In Library

08Short-term load forecasting based on multivariate time series prediction and weighted neural network with random weights and kernelsK. Lang, Mingyuan Zhang, Yongbo Yuan et al.2018 · Cluster Computing · 36 citations

09Metaheuristic design of feedforward neural networks: A review of two decades of researchVarun Ojha, A. Abraham, V. Snás̃el2017 · Engineering applications of artificial intelligence · 549 citations In Library

10Adam: A Method for Stochastic OptimizationDiederik P. Kingma, Jimmy Ba2014 · International Conference on Learning Representations · 169k citations In Library

11Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine TranslationKyunghyun Cho, B. V. Merrienboer, Çaglar Gülçehre et al.2014 · Conference on Empirical Methods in Natural Language Processing · 27k citations In Library

12On the difficulty of training recurrent neural networksRazvan Pascanu, Tomas Mikolov, Yoshua Bengio2012 · International Conference on Machine Learning · 6.0k citations In Library

Scroll for more · 2 remaining

Similar papers

Random error sampling-based recurrent neural network architecture optimizationAndrés Camero, J. Toutouh, E. Alba2019 · Engineering applications of artificial intelligence · 25 citations In Library

Random Error Sampling-based Recurrent Neural Network Architecture OptimizationAndrés Camero, J. Toutouh, E. Alba2019 · arXiv.org · 16 citations In Library

Evolutionary Recurrent Neural Architecture SearchShuo Tian, Kai Hu, Shasha Guo et al.2020 · IEEE Embedded Systems Letters · 4 citations In Library

Reinforced Evolutionary Neural Architecture SearchYukang Chen, Qian Zhang, Chang Huang et al.2018 · arXiv.org · 27 citations In Library

RetNAS: Reinforce Evolutionary Trend in Lightweight Neural Architecture SearchMuhan Xiao, Shijian Li, Li Zhang et al.2024 In Library

Multi-Objective Evolutionary Neural Architecture Search for Recurrent Neural NetworksReinhard Booysen, Anna Sergeevna Bosman2024 · arXiv (Cornell University) · 6 citations In Library

Efficient Self-learning Evolutionary Neural Architecture SearchZhengzhong Qiu, Wei Bi, Dong Xu et al.2023 · Applied Soft Computing In Library

RENAS: Reinforced Evolutionary Neural Architecture SearchYukang Chen, Gaofeng Meng, Qian Zhang et al.2019 · Computer Vision and Pattern Recognition · 126 citations In Library

Sampled Training and Node Inheritance for Fast Evolutionary Neural Architecture SearchHaoyu Zhang, Yaochu Jin, Ran Cheng et al.2020 · arXiv (Cornell University) · 11 citations In Library

Re-visiting Reservoir Computing architectures optimized by Evolutionary AlgorithmsSebastián Basterrech, Tarun Sharma2022 · arXiv (Cornell University) · 2 citations In Library

Source of record

arXiv

44B

Welcome.

This panel sticks with you. Pick anything below and it opens right beside it, so you can dig through 60,000-plus records without ever losing your spot here.

Travel 44B

LibraryPapers, policy, standards, statute LabsEvery organization building AI ModelsIntelligence, price, and speed BenchmarksThe evaluation catalog IncidentsAI failures and harms SearchOne field across everything DashboardThe state of AI accountability in NY 44B RegistryThe Art. 44-B compliance portal