Structural Self-Adaptation for Decentralized Pervasive Intelligence

Nikolic, Jovan; Pournaras, Evangelos

doi:10.1109/dsd.2019.00086

Cited by 6 publications

(11 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With this backpropagation mechanism, agents achieve a continuous selfimprovement until they find the optimal solution or they are trapped in a local minima. Strikingly earlier optimality results confirm the top 3% solution discovery under a monotonically decreasing inefficiency cost [5], [31]. I-EPOS also shows superior cost-effectiveness compared to related combinatorial optimization approaches [32].…”

Section: Human-centered Collective Learningmentioning

confidence: 61%

Section: Research Evolution and Milestonesmentioning

confidence: 99%

“…Several new insights are gained by studying the role of structure in collective learning: agents' placements in the tree structure that improve efficiency are determined as well as the most effective plan features based on which such placements can be made [31]. New large-scale optimality benchmarks are also introduced [31], [37]. Furthermore, the resilience of collective learning in distributed environments with uncertainties is studied, with findings showing that localizing collective learning in clustered branches as a result of agent failures can mitigate and even often boost the learning performance, i.e.…”

Section: Research Evolution and Milestonesmentioning

confidence: 99%

“…The optimality, complexity and scalability of collective learning are profiled in depth[5].Along with the introduction of collective learning, the applicability of this research expanded significantly beyond energy[5]. This expansion continues with new disruptive application scenarios: sharing economies (bike sharing), charging control of electric vehicles[38], traffic flow optimization[39], edge-to-cloud load-balancing and even an art project about the sonification of collective learning as an alternative means for general public to conceive such complex systems.Several new insights are gained by studying the role of structure in collective learning: agents' placements in the tree structure that improve efficiency are determined as well as the most effective plan features based on which such placements can be made[31]. New large-scale optimality benchmarks are also introduced[31],[37].…”

mentioning

confidence: 99%

“…This expansion continues with new disruptive application scenarios: sharing economies (bike sharing), charging control of electric vehicles[38], traffic flow optimization[39], edge-to-cloud load-balancing and even an art project about the sonification of collective learning as an alternative means for general public to conceive such complex systems.Several new insights are gained by studying the role of structure in collective learning: agents' placements in the tree structure that improve efficiency are determined as well as the most effective plan features based on which such placements can be made[31]. New large-scale optimality benchmarks are also introduced[31],[37]. Furthermore, the resilience of collective learning in distributed environments with uncertainties is studied, with findings showing that localizing collective learning in clustered branches as a result of agent failures can mitigate and even often boost the learning performance, i.e.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Collective Learning: A 10-Year Odyssey to Human-centered Distributed Intelligence

Pournaras

2020

2020 IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS)

Self Cite

View full text Add to dashboard Cite

This paper illustrates a 10-year research endeavor on collective learning, a paradigm for tackling tragedy of the commons problems in socio-technical systems using human-centered distributed intelligence. In contrast to mainstream centralized artificial intelligence (AI) allowing algorithmic discrimination and manipulative nudging, the decentralized approach of collective learning is by-design participatory and value-sensitive: it aligns with privacy, autonomy, fairness and democratic values. Engineering such values in a socio-technical system results in computational constraints that turn collective decision-making into complex combinatorial NP-hard problems. These are the problems that collective learning and the EPOS research project tackles. Collective learning finds striking applicability in energy, traffic, supply-chain and the self-management of sharing economies. This grand applicability and the social impact are demonstrated in this paper along with a future perspective of the collective learning paradigm.

show abstract

Section: Human-centered Collective Learningmentioning

confidence: 61%

Section: Research Evolution and Milestonesmentioning

confidence: 99%

Section: Research Evolution and Milestonesmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Collective Learning: A 10-Year Odyssey to Human-centered Distributed Intelligence

Pournaras

2020

2020 IEEE International Conference on Autonomic Computing and Self-Organizing Systems (ACSOS)

Self Cite

View full text Add to dashboard Cite

show abstract

Appliance-Level Flexible Scheduling for Socio-Technical Smart Grid Optimization

Fanitabasi

Pournaras

2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

Participation in residential energy demand response programs requires an active role by consumers. They contribute flexibility in how they use their appliances as the means to adjust energy consumption, and reduce demand peaks, possibly at the expense of their own comfort (e.g., thermal). Understanding the collective potential of appliance-level flexibility for reducing demand peaks is challenging and complex. For instance, physical characteristics of appliances, usage preferences, and comfort requirements all influence consumer flexibility, adoption, and effectiveness of demand response programs. To capture and study such socio-technical factors and trade-offs, this paper contributes a novel appliance-level flexible scheduling framework based on consumers' self-determined flexibility and comfort requirements. By utilizing this framework, this paper studies (i) consumers' usage preferences across various appliances, as well as their voluntary contribution of flexibility and willingness to sacrifice comfort for improving grid stability, (ii) impact of individual appliances on the collective goal of reducing demand peaks, and (iii) the effect of variable levels of flexibility, cooperation, and participation on the outcome of coordinated appliance scheduling. Experimental evaluation using a novel dataset collected via a smartphone app shows that higher consumer flexibility can significantly reduce demand peaks, with the oven having the highest system-wide potential for this. Overall, the cooperative approach allows for higher peak-shaving compared to non-cooperative schemes that focus entirely on the efficiency of individual appliances. The findings of this study can be used to design more cost-effective and granular (appliance-level) demand response programs in participatory and decentralized Smart Grids.

show abstract

Decentralized Edge-to-Cloud Load Balancing: Service Placement for the Internet of Things

et al. 2021

Self Cite

View full text Add to dashboard Cite

The Internet of Things (IoT) requires a new processing paradigm that inherits the scalability of the cloud while minimizing network latency using resources closer to the network edge. On the one hand, building up such flexibility within the edge-to-cloud continuum consisting of a distributed networked ecosystem of heterogeneous computing resources is challenging. On the other hand, IoT traffic dynamics and the rising demand for low-latency services foster the need for minimizing the response time and a balanced service placement. Load-balancing for fog computing becomes a cornerstone for cost-effective system management and operations. This paper studies two optimization objectives and formulates a decentralized load-balancing problem for IoT service placement: (global) IoT workload balance and (local) quality of service (QoS), in terms of minimizing the cost of deadline violation, service deployment, and unhosted services. The proposed solution, EPOS Fog, introduces a decentralized multi-agent system for collective learning that utilizes edge-to-cloud nodes to jointly balance the input workload across the network and minimize the costs involved in service execution. The agents locally generate possible assignments of requests to resources and then cooperatively select an assignment such that their combination maximizes edge utilization while minimizes service execution cost. Extensive experimental evaluation with realistic Google cluster workloads on various networks demonstrates the superior performance of EPOS Fog in terms of workload balance and QoS, compared to approaches such as First Fit and exclusively Cloud-based. The results confirm that EPOS Fog reduces service execution delay up to 25% and the load-balance of network nodes up to 90%. The findings also demonstrate how distributed computational resources on the edge can be utilized more cost-effectively by harvesting collective intelligence.INDEX TERMS Agent, cloud computing, collective learning, distributed optimization, edge computing, fog computing, internet of things (IoT), load-balancing, service placement.

show abstract

Structural Self-Adaptation for Decentralized Pervasive Intelligence

Cited by 6 publications

References 31 publications

Collective Learning: A 10-Year Odyssey to Human-centered Distributed Intelligence

Collective Learning: A 10-Year Odyssey to Human-centered Distributed Intelligence

Appliance-Level Flexible Scheduling for Socio-Technical Smart Grid Optimization

Decentralized Edge-to-Cloud Load Balancing: Service Placement for the Internet of Things

Contact Info

Product

Resources

About