2024/05/07 by Gergely Flamich, Flamich, Gergely, Lennie Wells +1
Computer Science · Decision Sciences · Engineering · #68P30 #Advanced Wireless Communication Techniques #E.4 #FOS: Computer and information sciences #G.3 #Information Theory (cs.IT) #Machine Learning (cs.LG) #Numerical Methods and Algorithms #Simulation Techniques and Applications
paper · pdf · doi:10.48550/arxiv.2405.04363
openalex publication_date 2024/05/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Channel simulation algorithms can efficiently encode random samples from a prescribed target distribution Q and find applications in machine learning-based lossy data compression. However, algorithms that encode exact samples usually have random runtime, limiting their applicability when a consistent encoding time is desirable. Thus, this paper considers approximate schemes with a fixed runtime instead. First, we strengthen a result of Agustsson and Theis and show that there is a class of pairs of target distribution Q and coding distribution P, for which the runtime of any approximate scheme scales at least super-polynomially in D_∞[Q \Vert P]. We then show, by contrast, that if we have access to an unnormalised Radon-Nikodym derivative r ∝ dQ/dP and knowledge of DKL[Q \Vert P], we can exploit global-bound, depth-limited A* coding to ensure TV[Q \Vert P] ≤ ε and maintain optimal coding performance with a sample complexity of only exp2((DKL[Q \Vert P] + o(1)) / ε).