Evolutionary Deep Learning for Car Park Occupancy Prediction in Smart Cities

By Andrés Camero, Jamal Toutouh, Daniel Stolfi, Enrique Alba. 2018

102 citations recorded by the source corpus.

Subfields: Building and Construction, Transportation

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Building and Construction

1. Andrés Camero
2. Enrique Alba
3. Jamal Toutouh
4. Javier Ferrer

In Transportation

1. Enrique Alba
2. Andrés Camero
3. Jamal Toutouh
4. Javier Ferrer

Citations in the corpus

• Bayesian Neural Architecture Search using A Training-Free Performance Metric (2020, background): “Hidden layers (M) [1,8] [1,3] [1,8] Look back (T) [2,30] [2,30] [2,30] Neurons per layer (N) [10,100] [1,100] [1,300]”
• Waste generation prediction under uncertainty in smart cities through deep neuroevolution (2019, methodology): “Nevertheless, the use of such a methodology in the domain of DL is still limited [14].”
• DLOPT: Deep Learning Optimization Library (2018, methodology): “Most of the hyper-parameter optimization techniques included in DLOPT are originals and are publicly available, i.e. the source code and several scientific publications 11,15,16 .”
• BIN-CT: Urban Waste Collection based in Predicting the Container Fill Level (2018, methodology): “However, nowadays the recurring neural network technique has emerged again as a good method for the generation of time series predictions (LeCun et al., 2015, Camero et al., 2018).”
• A Specialized Evolutionary Strategy Using Mean Absolute Error Random Sampling to Design Recurrent Neural Networks (2019, methodology): “[7] applied GA to search for the most efficient ones to improve the accuracy and the performance”
• Low-Cost Recurrent Neural Network Expected Performance Evaluation (2018, background): “In order to improve the DNNs learning process, a promising line of research bears on optimizing (modifying and adapting) the hyper-parameters that define the neural network architecture to the specific datasets by changing the activation functions, increasing or reducing the number of hidden layers,”
• A Survey on Evolutionary Neural Architecture Search (2020, methodology): “Different from focusing on the number of neurons and the number of layers in the feedforward neural network, some ENAS methods concern about how many times the RNN should be unfold [45], [46].”
• Random error sampling-based recurrent neural network architecture optimization (2019, methodology): “In [7] applied GA to search for the most e ffi cient ones to improve the accuracy and the performance regarding the most commonly used RNNs configurations.”