?url_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Adc&rft.title=An+Achievable+and+Analytic+Solution+to+Information+Bottleneck+for+Gaussian+Mixtures&rft.creator=Song%2C+Y&rft.creator=Wan%2C+K&rft.creator=Liao%2C+Z&rft.creator=Xu%2C+H&rft.creator=Caire%2C+G&rft.creator=Shamai%2C+S&rft.description=In+this+paper%2C+we+consider+a+remote+source+coding+problem+with+binary+phase+shift+keying+(BPSK)+modulation+sources%2C+where+observations+are+corrupted+by+additive+white+Gaussian+noise+(AWGN).+An+intermediate+node%2C+such+as+a+relay%2C+receives+these+observations+and+performs+further+compression+to+find+the+optimal+trade-off+between+complexity+and+relevance.+This+problem+can+be+formulated+as+an+information+bottleneck+(IB)+problem+with+Bernoulli+sources+and+Gaussian+mixture+observations%2C+for+which+no+closed-form+solution+is+known.+To+address+this+challenge%2C+we+propose+a+unified+achievable+scheme+that+employs+three+different+compression+strategies+for+intermediate+node+processing%2C+i.e.%2C+two-level+quantization%2C+multi-level+deterministic+quantization%2C+and+soft+quantization+with+tanh+function.+Comparative+analyses+with+existing+methods%2C+such+as+the+Blahut-Arimoto+(BA)+algorithm+and+the+Information+Dropout+approach%2C+are+performed+through+numerical+evaluations.+The+proposed+analytic+scheme+is+observed+to+consistently+approach+the+(numerically)+optimal+performance+over+a+range+of+signal-to-noise+ratios+(SNRs)%2C+confirming+its+effectiveness+in+the+considered+setting.&rft.subject=Quantization+(signal)%2C+%0D%0AClosed-form+solutions%2C+%0D%0AAWGN%2C+%0D%0APhase+modulation%2C+%0D%0ASource+coding%2C+%0D%0AApproximation+algorithms%2C+%0D%0ABinary+phase+shift+keying&rft.publisher=IEEE&rft.date=2024-08-19&rft.type=Proceedings+paper&rft.language=eng&rft.source=+++++In%3A++IEEE+International+Symposium+on+Information+Theory+-+Proceedings.++(pp.+pp.+2460-2465).++IEEE%3A+Athens%2C+Greece.+(2024)+++++&rft.format=text&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10198390%2F1%2FISIT_2023-YI_version_final_submission.pdf&rft.identifier=https%3A%2F%2Fdiscovery.ucl.ac.uk%2Fid%2Feprint%2F10198390%2F&rft.rights=open