In distributional semantic accounts of the meaning of noun-noun compounds
(e.g. starfish, bank account, houseboat) the important role of constituent
polysemy remains largely unaddressed(cf. the meaning of star in starfish vs.
star cluster vs. star athlete). Instead of semantic vectors that average over
the different meanings of a constituent, disambiguated vectors of the
constituents would be needed in order to see what these more specific
constituent meanings contribute to the meaning of the compound as a whole. This
paper presents a novel approach to this specific problem of word sense
disambiguation: set expansion. We build on the approach developed by Mahabal et
al. (2018) which was originally designed to solve the analogy problem. We
modified their method in such a way that it can address the problem of sense
disambiguation of compound constituents. The results of experiments with a data
set of almost 9000 compounds (LADEC, Gagné et al. 2019) suggest that this
approach is successful, yet the success is sensitive to the frequency with
which the compounds are attested.

本文介绍了一种新颖的 Word Sense Disambiguation 方法，基于 set expansion 的思想，旨在解决复合词中多义词汇的歧义问题，实验表明该方法具有一定的成功率，但对复合词出现频率敏感。

复合词构成成分的意义消岐

Sense disambiguation of compound constituents

Given a small set of seed entities (e.g., ``USA'', ``Russia''), corpus-based
set expansion is to induce an extensive set of entities which share the same
semantic class (Country in this example) from a given corpus. Set expansion
benefits a wide range of downstream applications in knowledge discovery, such
as web search, taxonomy construction, and query suggestion. Existing
corpus-based set expansion algorithms typically bootstrap the given seeds by
incorporating lexical patterns and distributional similarity. However, due to
no negative sets provided explicitly, these methods suffer from semantic drift
caused by expanding the seed set freely without guidance. We propose a new
framework, Set-CoExpan, that automatically generates auxiliary sets as negative
sets that are closely related to the target set of user's interest, and then
performs multiple sets co-expansion that extracts discriminative features by
comparing target set with auxiliary sets, to form multiple cohesive sets that
are distinctive from one another, thus resolving the semantic drift issue. In
this paper we demonstrate that by generating auxiliary sets, we can guide the
expansion process of target set to avoid touching those ambiguous areas around
the border with auxiliary sets, and we show that Set-CoExpan outperforms strong
baseline methods significantly.

本文提出了一种新的框架 Set-CoExpan，它可以自动生成辅助负集作为联系紧密的与用户所需目标集相关的负向集，通过将目标集与辅助集进行比较提取区分特征，形成多个互不相同的相互关联的集合，从而解决语义漂移问题。实验证明，通过生成辅助集，可以引导目标集的扩展过程，从而避免触及边界周围的模糊区域，同时 Set-CoExpan 可以显著优于强基线方法。

基于语料库及辅助集生成与协同扩展的集合扩展引导

Guiding Corpus-based Set Expansion by Auxiliary Sets Generation and  Co-Expansion

We study the problem of designing models for machine learning tasks defined
on \emph{sets}. In contrast to traditional approach of operating on fixed
dimensional vectors, we consider objective functions defined on sets that are
invariant to permutations. Such problems are widespread, ranging from
estimation of population statistics \cite{poczos13aistats}, to anomaly
detection in piezometer data of embankment dams \cite{Jung15Exploration}, to
cosmology \cite{Ntampaka16Dynamical,Ravanbakhsh16ICML1}. Our main theorem
characterizes the permutation invariant functions and provides a family of
functions to which any permutation invariant objective function must belong.
This family of functions has a special structure which enables us to design a
deep network architecture that can operate on sets and which can be deployed on
a variety of scenarios including both unsupervised and supervised learning
tasks. We also derive the necessary and sufficient conditions for permutation
equivariance in deep models. We demonstrate the applicability of our method on
population statistic estimation, point cloud classification, set expansion, and
outlier detection.

研究设计用于基于集合的机器学习任务模型的问题，提出一种特殊结构的函数族，设计可以在未监督和监督学习任务中应用的深度网络架构，并演示了在人口统计估计，点云分类，集扩展和异常检测等任务上的应用。