Matches in SemOpenAlex for { <https://semopenalex.org/work/W4307321791> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4307321791 abstract "We study the problem of covering and learning sums $X = X_1 + cdots + X_n$ of independent integer-valued random variables $X_i$ (SIIRVs) with unbounded, or even infinite, support. De et al. at FOCS 2018, showed that the maximum value of the collective support of $X_i$'s necessarily appears in the sample complexity of learning $X$. In this work, we address two questions: (i) Are there general families of SIIRVs with unbounded support that can be learned with sample complexity independent of both $n$ and the maximal element of the support? (ii) Are there general families of SIIRVs with unbounded support that admit proper sparse covers in total variation distance? As for question (i), we provide a set of simple conditions that allow the unbounded SIIRV to be learned with complexity $text{poly}(1/epsilon)$ bypassing the aforementioned lower bound. We further address question (ii) in the general setting where each variable $X_i$ has unimodal probability mass function and is a different member of some, possibly multi-parameter, exponential family $mathcal{E}$ that satisfies some structural properties. These properties allow $mathcal{E}$ to contain heavy tailed and non log-concave distributions. Moreover, we show that for every $epsilon > 0$, and every $k$-parameter family $mathcal{E}$ that satisfies some structural assumptions, there exists an algorithm with $tilde{O}(k) cdot text{poly}(1/epsilon)$ samples that learns a sum of $n$ arbitrary members of $mathcal{E}$ within $epsilon$ in TV distance. The output of the learning algorithm is also a sum of random variables whose distribution lies in the family $mathcal{E}$. En route, we prove that any discrete unimodal exponential family with bounded constant-degree central moments can be approximated by the family corresponding to a bounded subset of the initial (unbounded) parameter space." @default.
- W4307321791 created "2022-10-31" @default.
- W4307321791 creator A5042852153 @default.
- W4307321791 creator A5083461583 @default.
- W4307321791 creator A5090925161 @default.
- W4307321791 date "2022-10-24" @default.
- W4307321791 modified "2023-09-27" @default.
- W4307321791 title "Learning and Covering Sums of Independent Random Variables with Unbounded Support" @default.
- W4307321791 doi "https://doi.org/10.48550/arxiv.2210.13313" @default.
- W4307321791 hasPublicationYear "2022" @default.
- W4307321791 type Work @default.
- W4307321791 citedByCount "0" @default.
- W4307321791 crossrefType "posted-content" @default.
- W4307321791 hasAuthorship W4307321791A5042852153 @default.
- W4307321791 hasAuthorship W4307321791A5083461583 @default.
- W4307321791 hasAuthorship W4307321791A5090925161 @default.
- W4307321791 hasBestOaLocation W43073217911 @default.
- W4307321791 hasConcept C105795698 @default.
- W4307321791 hasConcept C111472728 @default.
- W4307321791 hasConcept C114614502 @default.
- W4307321791 hasConcept C118615104 @default.
- W4307321791 hasConcept C122123141 @default.
- W4307321791 hasConcept C134306372 @default.
- W4307321791 hasConcept C138885662 @default.
- W4307321791 hasConcept C14036430 @default.
- W4307321791 hasConcept C151376022 @default.
- W4307321791 hasConcept C154945302 @default.
- W4307321791 hasConcept C177264268 @default.
- W4307321791 hasConcept C17744445 @default.
- W4307321791 hasConcept C182365436 @default.
- W4307321791 hasConcept C18903297 @default.
- W4307321791 hasConcept C199360897 @default.
- W4307321791 hasConcept C199539241 @default.
- W4307321791 hasConcept C200288055 @default.
- W4307321791 hasConcept C2777299769 @default.
- W4307321791 hasConcept C2778445095 @default.
- W4307321791 hasConcept C2780586882 @default.
- W4307321791 hasConcept C33923547 @default.
- W4307321791 hasConcept C41008148 @default.
- W4307321791 hasConcept C55974624 @default.
- W4307321791 hasConcept C78458016 @default.
- W4307321791 hasConcept C86803240 @default.
- W4307321791 hasConcept C97137487 @default.
- W4307321791 hasConceptScore W4307321791C105795698 @default.
- W4307321791 hasConceptScore W4307321791C111472728 @default.
- W4307321791 hasConceptScore W4307321791C114614502 @default.
- W4307321791 hasConceptScore W4307321791C118615104 @default.
- W4307321791 hasConceptScore W4307321791C122123141 @default.
- W4307321791 hasConceptScore W4307321791C134306372 @default.
- W4307321791 hasConceptScore W4307321791C138885662 @default.
- W4307321791 hasConceptScore W4307321791C14036430 @default.
- W4307321791 hasConceptScore W4307321791C151376022 @default.
- W4307321791 hasConceptScore W4307321791C154945302 @default.
- W4307321791 hasConceptScore W4307321791C177264268 @default.
- W4307321791 hasConceptScore W4307321791C17744445 @default.
- W4307321791 hasConceptScore W4307321791C182365436 @default.
- W4307321791 hasConceptScore W4307321791C18903297 @default.
- W4307321791 hasConceptScore W4307321791C199360897 @default.
- W4307321791 hasConceptScore W4307321791C199539241 @default.
- W4307321791 hasConceptScore W4307321791C200288055 @default.
- W4307321791 hasConceptScore W4307321791C2777299769 @default.
- W4307321791 hasConceptScore W4307321791C2778445095 @default.
- W4307321791 hasConceptScore W4307321791C2780586882 @default.
- W4307321791 hasConceptScore W4307321791C33923547 @default.
- W4307321791 hasConceptScore W4307321791C41008148 @default.
- W4307321791 hasConceptScore W4307321791C55974624 @default.
- W4307321791 hasConceptScore W4307321791C78458016 @default.
- W4307321791 hasConceptScore W4307321791C86803240 @default.
- W4307321791 hasConceptScore W4307321791C97137487 @default.
- W4307321791 hasLocation W43073217911 @default.
- W4307321791 hasOpenAccess W4307321791 @default.
- W4307321791 hasPrimaryLocation W43073217911 @default.
- W4307321791 hasRelatedWork W1971068802 @default.
- W4307321791 hasRelatedWork W2007566017 @default.
- W4307321791 hasRelatedWork W2044035340 @default.
- W4307321791 hasRelatedWork W2078141285 @default.
- W4307321791 hasRelatedWork W2078932748 @default.
- W4307321791 hasRelatedWork W2100199806 @default.
- W4307321791 hasRelatedWork W2371742721 @default.
- W4307321791 hasRelatedWork W2386005976 @default.
- W4307321791 hasRelatedWork W2776018879 @default.
- W4307321791 hasRelatedWork W2999954399 @default.
- W4307321791 isParatext "false" @default.
- W4307321791 isRetracted "false" @default.
- W4307321791 workType "article" @default.