Abstract. Learning latent representations is playing a pivotal role in machine learning and many application areas. Previous work on relational topic models (RTM) has shown promise on learning latent topical representations for describing relational document networks and predicting pairwise links. However under a probabilistic formulation with normalization constraints, RTM could be ineffective in controlling the sparsity of the topical representations, and may often need to make strict meanfield assumptions for approximate inference. This paper presents sparse relational topic models (SRTM) under a non-probabilistic formulation that can effectively control the sparsity via a sparsity-inducing regularizer. Our model can also handle imbalance issues in real networks via introducing various cost parameters for positive and negative links. The deterministic optimization problem of SRTM admits efficient coordinate descent algorithms. We also present a generalization to consider all pairwise topic interactions. Our empirical results on several real network datasets demonstrate better performance on link prediction, sparser latent representations, and faster running time than the competitors under a probabilistic formulation.

In this paper, we propose a fully supervised speaker diarization approach, named unbounded interleaved-state recurrent neural networks (UIS-RNN). Given extracted speaker-discriminative embeddings (a.k.a. d-vectors) from input utterances, each individual speaker is modeled by a parameter-sharing RNN, while the RNN states for different speakers interleave in the time domain. This RNN is naturally integrated with a distance-dependent Chinese restaurant process (ddCRP) to accommodate an unknown number of speakers. Our system is fully supervised and is able to learn from examples where time-stamped speaker labels are annotated. We achieved a 7.6% diarization error rate on NIST SRE 2000 CALLHOME, which is better than the state-of-the-art method using spectral clustering. Moreover, our method decodes in an online fashion while most state-of-the-art systems rely on offline clustering.

Contextuality is considered as an intrinsic signature of non-classicality, and a crucial resource for achieving unique advantages of quantum information processing. However, recently there have been debates on whether classical fields may also demonstrate contextuality. Here we experimentally configure a contextuality test for optical fields, adopting various definitions of measurement events, and analyse how the definitions affect the emergence of non-classical correlations. The heralded single photon state, a typical non-classical light field, manifests contextuality in our setup, while contextuality for classical coherent fields strongly depends on the specific definition of measurement events which is equivalent to filtering the non-classical component of the input state. Our results highlight the importance of definition of measurement events to demonstrate contextuality, and link the contextual correlations to non-classicality defined by quasi-probabilities in phase space. arXiv:1810.07966v2 [quant-ph]

No abstract