Comprehensive Review of Deep Reinforcement Learning Methods and Applications in Economics

Mosavi, Amir; Ghamisi, Pedram; Faghan, Yaser; Duan, Puhong; Ardabili, Sina Faizollahzadeh; Salwana, Ely; Band, Shahab S.

doi:10.2139/ssrn.3711731

Cited by 9 publications

(8 citation statements)

References 78 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many modifications to the base DQN model have been proposed and both [22] and [23] review this literature further. Of interest herein is Dueling DQN (DDQN), which look at both the value of being in a particular state and the advantage of a given action vs a different action [24].…”

Section: Value-based Methodsmentioning

confidence: 99%

Assessing Multi-Agent Reinforcement Learning Algorithms for Autonomous Sensor Resource Management

Bihl¹,

Jones²,

Farr³

et al. 2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

Unmanned aerial vehicles (UAVs) have applications in search and rescue operations and such operations could be more efficient by using appropriate artificial intelligence (AI) to enable a UAV agent to operate autonomously. Sensor resource management (SRM), which leverages capabilities across location intelligence, facilitates the efficient and effective use of UAVs and their sensors to complete a set of tasks. Furthermore, multiple UAVs, each with different sensor configurations, must be considered when maximizing mission effects. Instantiating operational autonomy for such teams requires considerable coordination. One AI approach relevant to this task is multi-agent reinforcement learning (MARL). However, MARL has seen limited prior use in SRM. This work evaluates the trade-space of MARL algorithms with respect to performing heterogeneous sensor resource management (SRM) tasks, considers the concept of evaluating MARL in a test and evaluation framework and compares a suit of algorithms with random and Bayesian hyperparameter optimization methods.

show abstract

Section: Value-based Methodsmentioning

confidence: 99%

Assessing Multi-Agent Reinforcement Learning Algorithms for Autonomous Sensor Resource Management

Bihl¹,

Jones²,

Farr³

et al. 2022

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

show abstract

“…11 presents the architecture of CNN technique. Additional explanations about RNN method have been presented in our previous study entitled "list of DL techniques" (Mosavi, Ardabili et al 2019). (Despotovic, Koch et al 2019) an exploration for the of heating demand in building sector using CNN technique.…”

Section: Convolutional Neural Network (Cnn) Based Studiesmentioning

confidence: 99%

Systematic review of deep learning and machine learning for building energy

Ardabili¹,

Abdilalizadeh²,

Makó³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

The building energy (BE) management has an essential role in urban sustainability and smart cities. Recently, the novel data science and data-driven technologies have shown significant progress in analyzing the energy consumption and energy demand data sets for a smarter energy management. The machine learning (ML) and deep learning (DL) methods and applications, in particular, have been promising for the advancement of the accurate and high-performance energy models. The present study provides a comprehensive review of ML and DL-based techniques applied for handling BE systems, and it further evaluates the performance of these techniques. Through a systematic review and a comprehensive taxonomy, the advances of ML and DL-based techniques are carefully investigated, and the promising models are introduced. According to the results obtained for energy demand forecasting, the hybrid and ensemble methods are located in high robustness range, SVM-based methods are located in good robustness limitation, ANN-based methods are located in medium robustness limitation and linear regression models are located in low robustness limitations. On the other hand, for energy consumption forecasting, DL-based, hybrid, and ensemble-based models provided the highest robustness score. ANN, SVM, and single ML models provided good and medium robustness and LR-based models provided the lower robustness score. In addition, for energy load forecasting, LR-based models provided the lower robustness score. The hybrid and ensemble-based models provided a higher robustness score. The DL-based and SVM-based techniques provided a good robustness score and ANN-based techniques provided a medium robustness score.

show abstract

“…These (learnable) filters are sparse and shared across the entire input and are adjusted during the training process, in order to be activated when certain features are detected. Thus, we can define in Equation ( 2) the output result/ feature map y k , f j being the chosen activation function [41].…”

Section: Convolutional Neural Networkmentioning

confidence: 99%

Ocean wave power forecasting using convolutional neural networks

Bento

Pombo

Calado

et al. 2021

IET Renewable Power Gen

View full text Add to dashboard Cite

Climate change "fuelled" by anthropogenic causes has been identified as the greatest threat faced by societies. In this respect, the roadmap to a "greener" generation mix certainly includes a greater heterogeneity in terms of renewable energy sources. In this regard, one of the leading candidates is ocean wave energy. One of the issues with renewables in general is their unpredictably and variability, as it is crucial to address the subject of wave power forecasting, to facilitate a future market integration. Hence, to tackle this prediction problem, a new approach to short-term wave power forecasting is proposed, based on deep learning capabilities. These highly popular networks were traditionally developed to deal with images (2D data), so the authors discuss all the necessary implementation and design details to employ these networks with 1D input data, to solve a regression-based problem. These case-studies include wave data from three different locations. The proposed approach was tested across all seasons of the year, revealing the suitability to extract the relevant input data dependencies from the time-series. As such, especially for horizons up to 6 h, the proposed approach outperforms other conventional methods.

show abstract

Comprehensive Review of Deep Reinforcement Learning Methods and Applications in Economics

Cited by 9 publications

References 78 publications

Assessing Multi-Agent Reinforcement Learning Algorithms for Autonomous Sensor Resource Management

Assessing Multi-Agent Reinforcement Learning Algorithms for Autonomous Sensor Resource Management

Systematic review of deep learning and machine learning for building energy

Ocean wave power forecasting using convolutional neural networks

Contact Info

Product

Resources

About