Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381785934> ?p ?o ?g. }
- W4381785934 endingPage "2627" @default.
- W4381785934 startingPage "2609" @default.
- W4381785934 abstract "The emerging field semantic communication is driving the research of end-to-end data transmission. By utilizing the powerful representation ability of deep learning models, learned data transmission schemes have exhibited superior performance than the established source and channel coding methods. While, so far, research efforts mainly concentrated on architecture and model improvements toward a static target domain. Despite their successes, such learned models are still suboptimal due to the limitations in model capacity and imperfect optimization and generalization, particularly when the testing data distribution or channel response is different from that adopted for model training, as is likely to be the case in real-world. To tackle this, in this paper, we propose a novel online learned joint source and channel coding approach that leverages the deep learning model’s overfitting property. Specifically, we update the off-the-shelf pre-trained models after deployment in a lightweight online fashion to adapt to the distribution shifts in source data and environment domain. We take the overfitting concept to the extreme, proposing a series of implementation-friendly methods to adapt the codec model or representations to an individual data or channel state instance, which can further lead to substantial gains in terms of the end-to-end rate-distortion performance. Accordingly, the streaming ingredients include both the semantic representations of source data and the online updated decoder model parameters. The system design is formulated as a joint optimization problem whose goal is to minimize the loss function, a tripartite trade-off among the data stream bandwidth cost, model stream bandwidth cost, and end-to-end distortion. The proposed methods enable the communication-efficient adaptation for all parameters in the network without sacrificing decoding speed. Extensive experiments, including user study, on continually changing target source data and wireless channel environments, demonstrate the effectiveness and efficiency of our approach, on which we outperform existing state-of-the-art engineered transmission scheme (VVC combined with 5G LDPC coded transmission)." @default.
- W4381785934 created "2023-06-24" @default.
- W4381785934 creator A5001455682 @default.
- W4381785934 creator A5002669383 @default.
- W4381785934 creator A5008455605 @default.
- W4381785934 creator A5044461977 @default.
- W4381785934 creator A5055151897 @default.
- W4381785934 creator A5065170250 @default.
- W4381785934 creator A5069582226 @default.
- W4381785934 creator A5091040593 @default.
- W4381785934 date "2023-08-01" @default.
- W4381785934 modified "2023-10-15" @default.
- W4381785934 title "Toward Adaptive Semantic Communications: Efficient Data Transmission via Online Learned Nonlinear Transform Source-Channel Coding" @default.
- W4381785934 cites W1995875735 @default.
- W4381785934 cites W2027573699 @default.
- W4381785934 cites W2051888076 @default.
- W4381785934 cites W2072079297 @default.
- W4381785934 cites W2073750184 @default.
- W4381785934 cites W2119047110 @default.
- W4381785934 cites W2129652681 @default.
- W4381785934 cites W2134246836 @default.
- W4381785934 cites W2146395539 @default.
- W4381785934 cites W2175325787 @default.
- W4381785934 cites W2403576189 @default.
- W4381785934 cites W2566065221 @default.
- W4381785934 cites W2599701415 @default.
- W4381785934 cites W2741137940 @default.
- W4381785934 cites W2789254838 @default.
- W4381785934 cites W2792557392 @default.
- W4381785934 cites W2935381027 @default.
- W4381785934 cites W2946889564 @default.
- W4381785934 cites W2962785568 @default.
- W4381785934 cites W2963174256 @default.
- W4381785934 cites W2981613960 @default.
- W4381785934 cites W3016715915 @default.
- W4381785934 cites W3035047545 @default.
- W4381785934 cites W3036851434 @default.
- W4381785934 cites W3046761316 @default.
- W4381785934 cites W3082548248 @default.
- W4381785934 cites W3095497211 @default.
- W4381785934 cites W3102125291 @default.
- W4381785934 cites W3107617916 @default.
- W4381785934 cites W3110286842 @default.
- W4381785934 cites W3129163321 @default.
- W4381785934 cites W3135530072 @default.
- W4381785934 cites W3177376120 @default.
- W4381785934 cites W3203825070 @default.
- W4381785934 cites W3207494168 @default.
- W4381785934 cites W3213872105 @default.
- W4381785934 cites W3216777582 @default.
- W4381785934 cites W4210317917 @default.
- W4381785934 cites W4226042246 @default.
- W4381785934 cites W4229333757 @default.
- W4381785934 cites W4290713755 @default.
- W4381785934 cites W4312590925 @default.
- W4381785934 cites W4313350178 @default.
- W4381785934 cites W4313455370 @default.
- W4381785934 doi "https://doi.org/10.1109/jsac.2023.3288246" @default.
- W4381785934 hasPublicationYear "2023" @default.
- W4381785934 type Work @default.
- W4381785934 citedByCount "0" @default.
- W4381785934 crossrefType "journal-article" @default.
- W4381785934 hasAuthorship W4381785934A5001455682 @default.
- W4381785934 hasAuthorship W4381785934A5002669383 @default.
- W4381785934 hasAuthorship W4381785934A5008455605 @default.
- W4381785934 hasAuthorship W4381785934A5044461977 @default.
- W4381785934 hasAuthorship W4381785934A5055151897 @default.
- W4381785934 hasAuthorship W4381785934A5065170250 @default.
- W4381785934 hasAuthorship W4381785934A5069582226 @default.
- W4381785934 hasAuthorship W4381785934A5091040593 @default.
- W4381785934 hasBestOaLocation W43817859342 @default.
- W4381785934 hasConcept C108583219 @default.
- W4381785934 hasConcept C119857082 @default.
- W4381785934 hasConcept C154945302 @default.
- W4381785934 hasConcept C161765866 @default.
- W4381785934 hasConcept C22019652 @default.
- W4381785934 hasConcept C31258907 @default.
- W4381785934 hasConcept C41008148 @default.
- W4381785934 hasConcept C50644808 @default.
- W4381785934 hasConcept C557945733 @default.
- W4381785934 hasConcept C67186912 @default.
- W4381785934 hasConcept C76155785 @default.
- W4381785934 hasConcept C77088390 @default.
- W4381785934 hasConceptScore W4381785934C108583219 @default.
- W4381785934 hasConceptScore W4381785934C119857082 @default.
- W4381785934 hasConceptScore W4381785934C154945302 @default.
- W4381785934 hasConceptScore W4381785934C161765866 @default.
- W4381785934 hasConceptScore W4381785934C22019652 @default.
- W4381785934 hasConceptScore W4381785934C31258907 @default.
- W4381785934 hasConceptScore W4381785934C41008148 @default.
- W4381785934 hasConceptScore W4381785934C50644808 @default.
- W4381785934 hasConceptScore W4381785934C557945733 @default.
- W4381785934 hasConceptScore W4381785934C67186912 @default.
- W4381785934 hasConceptScore W4381785934C76155785 @default.
- W4381785934 hasConceptScore W4381785934C77088390 @default.
- W4381785934 hasFunder F4320321001 @default.
- W4381785934 hasFunder F4320321470 @default.
- W4381785934 hasFunder F4320322919 @default.