A Global Brain fuelled by Local intelligence: Optimizing Mobile Services and Networks with AI

Naas, Si‐Ahmed; Mohammed, Thaha; Sigg, Stephan

doi:10.1109/msn50589.2020.00021

Cited by 2 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The EG and EEG programs are solved by employing the CVXPY modeling language [30] and the MOSEK solver [31] after normalizing the capacity (i.e., the maximum number of layers) of the considered DNN architectures. The base demand vector is determined for a distributed intelligence scenario [19,20], in which the orchestrator predicts the model accuracy expected from the weights by using a gradient boosting machine with regression trees [22]. Simulations are carried out by using the independent replication method, in which randomly selected subsets of the data are assigned to different SPs.…”

Section: A Market Equilibriummentioning

confidence: 99%

See 1 more Smart Citation

Knowledge Sharing in AI Services: A Market-Based Approach

Mohammed

Naas

Sigg

et al. 2023

IEEE Internet Things J.

Self Cite

View full text Add to dashboard Cite

Today's deep neural networks (DNNs) are very accurate when trained on a large amount of data. However, suitable input might not be available or may require extensive data collection. Data sharing is one option to address these issues, but it is generally impractical because of privacy concerns or due to the problematic process of finding a sharing agreement. Instead, this work considers knowledge sharing by first exchanging the weights of pre-trained DNNs and then applying transfer learning. Specifically, it addresses the economics of knowledge sharing in AI services by taking a market-based approach. In detail, a model based on Fisher's market is devised for optimal knowledge sharing, defined as the gain in inference accuracy from exchanging DNN weights. The proposed approach is shown to reach a market equilibrium and to satisfy important economic properties, including Pareto optimality. A technique for weight fusion is also introduced to merge acquired knowledge with the existing one. Finally, an extensive evaluation is conducted in a distributed intelligence scenario. The obtained results show that the proposed solution is efficient and that weight fusion with transfer learning significantly increases inference accuracy compared to the original DNN, without the overhead of federated learning.

show abstract

Section: A Market Equilibriummentioning

confidence: 99%

“…Section IV will show how to derive such an accuracy for the case of distributed intelligence based on combined local / global knowledge[19,20].…”

mentioning

confidence: 99%