Skip Training for Multi-Agent Reinforcement Learning Controller for Industrial Wave Energy Converters

Sarkar, Soumyendu; Gundecha, Vineet; Ghorbanpour, Sahand; Shmakov, Alexander; Babu, Ashwin Ramesh; Pichard, Alexandre; Cocho, Mathieu

doi:10.1109/case49997.2022.9926561

Cited by 9 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CAM (Zhou et al 2016) localizes image regions that are most important for a target class. CAM has gained much popularity and promotes numerous further studies (Selvaraju et al 2017;Lee et al 2021;Hasany, Petitjean, and Mériaudeau 2023;Sarkar et al 2023). Notably, Grad-CAM (Selvaraju et al 2017) extends CAM by using gradients to weigh the contribution of each filter to the target class.…”

Section: Related Workmentioning

confidence: 99%

PICNN: A Pathway towards Interpretable Convolutional Neural Networks

Guo,

Yang,

Yin

et al. 2024

AAAI

View full text Add to dashboard Cite

Convolutional Neural Networks (CNNs) have exhibited great performance in discriminative feature learning for complex visual tasks. Besides discrimination power, interpretability is another important yet under-explored property for CNNs. One difficulty in the CNN interpretability is that filters and image classes are entangled. In this paper, we introduce a novel pathway to alleviate the entanglement between filters and image classes. The proposed pathway groups the filters in a late conv-layer of CNN into class-specific clusters. Clusters and classes are in a one-to-one relationship. Specifically, we use the Bernoulli sampling to generate the filter-cluster assignment matrix from a learnable filter-class correspondence matrix. To enable end-to-end optimization, we develop a novel reparameterization trick for handling the non-differentiable Bernoulli sampling. We evaluate the effectiveness of our method on ten widely used network architectures (including nine CNNs and a ViT) and five benchmark datasets. Experimental results have demonstrated that our method PICNN (the combination of standard CNNs with our proposed pathway) exhibits greater interpretability than standard CNNs while achieving higher or comparable discrimination power.

show abstract

Section: Related Workmentioning

confidence: 99%

PICNN: A Pathway towards Interpretable Convolutional Neural Networks

Guo,

Yang,

Yin

et al. 2024

AAAI

View full text Add to dashboard Cite

show abstract

“…Grid data, which includes grid energy and carbon distribution, weather conditions, and compute load, is configured to be queried from a database at every time step. On the other hand, the variables like DC temperature, IT Load, Unassigned flexible load, DC Energy, and Battery state of charge information are obtained via the exchange of information between the individual MDPs as a part of the Data Center Simulator which has the Load Shifting Model (Acun et al 2023), Energy Plus model for the data center thermodynamics (Crawley et al 2000) and Battery Storage Model (Acun et al 2023;Sarkar et al 2023b). The information exchange processes occur through the RL Interface with Open AI Gym wrappers.…”

Section: Inputmentioning

confidence: 99%

Carbon Footprint Reduction for Sustainable Data Centers in Real-Time

Sarkar,

Naug,

Luna

et al. 2024

AAAI

View full text Add to dashboard Cite

As machine learning workloads are significantly increasing energy consumption, sustainable data centers with low carbon emissions are becoming a top priority for governments and corporations worldwide. This requires a paradigm shift in optimizing power consumption in cooling and IT loads, shifting flexible loads based on the availability of renewable energy in the power grid, and leveraging battery storage from the uninterrupted power supply in data centers, using collaborative agents. The complex association between these optimization strategies and their dependencies on variable external factors like weather and the power grid carbon intensity makes this a hard problem. Currently, a real-time controller to optimize all these goals simultaneously in a dynamic real-world setting is lacking. We propose a Data Center Carbon Footprint Reduction (DC-CFR) multi-agent Reinforcement Learning (MARL) framework that optimizes data centers for the multiple objectives of carbon footprint reduction, energy consumption, and energy cost. The results show that the DC-CFR MARL agents effectively resolved the complex interdependencies in optimizing cooling, load shifting, and energy storage in real-time for various locations under real-world dynamic weather and grid carbon intensity conditions. DC-CFR significantly outperformed the industry-standard ASHRAE controller with a considerable reduction in carbon emissions (14.5%), energy usage (14.4%), and energy cost (13.7%) when evaluated over one year across multiple geographical regions.

show abstract

“…Supervised learning Classification ANN Coastal vulnerability map Ennouali et al, (2023); Coastal waters classification Pereira and Ebecken, (2009); Coastal Altimetric Waveforms Xu et al, (2021); Sea Surface Temperature Imagery Reggiannini et al, (2022) SVM RF K-Nearest neighbor Naive-Bayes classifer Regression ANN Wave condition James et al, (2018); Significant wave height Ali et al, (2023); Breaking wave height Duong et al, (2023); Sediment load Latif et al, (2023); Wave attenuation Kim et alReal-time control of coastal urban stormwater systemsBowes et al, (2022); Flood mitigationBowes et al, (2021); Maximize Energy EfficiencySarkar et al, (2022) …”

mentioning

confidence: 99%

Machine learning application in modelling marine and coastal phenomena: a critical review

Pourzangbar,

Jalali,

Brocchini

2023

Front. Environ. Eng.

View full text Add to dashboard Cite

This study provides an extensive review of over 200 journal papers focusing on Machine Learning (ML) algorithms’ use for promoting a sustainable management of the marine and coastal environments. The research covers various facets of ML algorithms, including data preprocessing and handling, modeling algorithms for distinct phenomena, model evaluation, and use of dynamic and integrated models. Given that machine learning modeling relies on experience or trial-and-error, examining previous applications in marine and coastal modeling is proven to be beneficial. The performance of different ML methods used to predict wave heights was analyzed to ascertain which method was superior with various datasets. The analysis of these papers revealed that properly developed ML methods could successfully be applied to multiple aspects. Areas of application include data collection and analysis, pollutant and sediment transport, image processing and deep learning, and identification of potential regions for aquaculture and wave energy activities. Additionally, ML methods aid in structural design and optimization and in the prediction and classification of oceanographic parameters. However, despite their potential advantages, dynamic and integrated ML models remain underutilized in marine projects. This research provides insights into ML’s application and invites future investigations to exploit ML’s untapped potential in marine and coastal sustainability.

show abstract

Skip Training for Multi-Agent Reinforcement Learning Controller for Industrial Wave Energy Converters

Cited by 9 publications

References 0 publications

PICNN: A Pathway towards Interpretable Convolutional Neural Networks

PICNN: A Pathway towards Interpretable Convolutional Neural Networks

Carbon Footprint Reduction for Sustainable Data Centers in Real-Time

Machine learning application in modelling marine and coastal phenomena: a critical review

Contact Info

Product

Resources

About