Afra Alishahi, Grzegorz Chrupała, Alejandrina Cristia et al.
We present the visually-grounded language modelling track that was introduced in the Zero-Resource Speech challenge, 2021 edition, 2nd round. We motivate the new track and discuss participation rules in detail. We also present the two baseline systems that were developed for this track.
Yu Chen, Lingfei Wu, Mohammed J. Zaki
When answering natural language questions over knowledge bases (KBs), different question components and KB aspects play different roles. However, most existing embedding-based methods for knowledge base question answering (KBQA) ignore the subtle inter-relationships between the question and the KB (e.g., entity types, relation paths and context). In this work, we propose to directly model the two-way flow of interactions between the questions and the KB via a novel Bidirectional Attentive Memory Network, called BAMnet. Requiring no external resources and only very few hand-crafted features, on the WebQuestions benchmark, our method significantly outperforms existing information-retrieval based methods, and remains competitive with (hand-crafted) semantic parsing based methods. Also, since we use attention mechanisms, our method offers better interpretability compared to other baselines.
Yu Chen, Xiufen Zou
Inference of gene regulatory networks (GRNs) based on experimental data is a challenging task in bioinformatics. In this paper, we present a bi-objective minimization model (BoMM) for inference of GRNs, where one objective is the fitting error of derivatives, and the other is the number of connections in the network. To solve the BoMM efficiently, we propose a multi-objective evolutionary algorithm (MOEA), and utilize the separable parameter estimation method (SPEM) decoupling the ordinary differential equation (ODE) system. Then, the Akaike Information Criterion (AIC) is employed to select one inference result from the obtained Pareto set. Taking the S-system as the investigated GRN model, our method can properly identify the topologies and parameter values of benchmark systems. There is no need to preset problem-dependent parameter values to obtain appropriate results, and thus, our method could be applicable to inference of various GRNs models.