We propose a novel data augmentation for labeled sentences called contextual augmentation. We assume an invariance that sentences are natural even if the words in the sentences are replaced with other words with paradigmatic relations. We stochastically replace words with other words that are predicted by a bi-directional language model at the word positions. Words predicted according to a context are numerous but appropriate for the augmentation of the original words. Furthermore, we retrofit a language model with a label-conditional architecture, which allows the model to augment sentences without breaking the label-compatibility. Through the experiments for six various different text classification tasks, we demonstrate that the proposed method improves classifiers based on the convolutional or recurrent neural networks.

Comprehension of spoken natural language is an essential skill for robots to communicate with humans effectively. However, handling unconstrained spoken instructions is challenging due to (1) complex structures and the wide variety of expressions used in spoken language, and (2) inherent ambiguity of human instructions. In this paper, we propose the first comprehensive system for controlling robots with unconstrained spoken language, which is able to effectively resolve ambiguity in spoken instructions. Specifically, we integrate deep learning-based object detection together with natural language processing technologies to handle unconstrained spoken instructions, and propose a method for robots to resolve instruction ambiguity through dialogue. Through our experiments on both a simulated environment as well as a physical industrial robot arm, we demonstrate the ability of our system to understand natural instructions from human operators effectively, and show how higher success rates of the object picking task can be achieved through an interactive clarification process. 1

Crystal lattice defects often degrade device functionality, but engineering these defects may have value in future electronic and magnetic device applications. For example, dislocations--one-dimensional lattice defects with locally distinct atomic-scale structures--exhibit unique and localized electrical properties and can be used as a template for producing conducting nanowires in insulating crystals. It has also been predicted that spin-polarized current may flow along dislocations in topological insulators. Although it is expected that the magnetic properties of dislocations will differ from those of the lattice, their fundamental characterization at the individual level has received little attention. Here, we demonstrate that dislocations in NiO crystals show unique magnetic properties. Magnetic force microscopy imaging clearly reveals ferromagnetic ordering of individual dislocations in antiferromagnetic NiO, originating from the local non-stoichiometry of the dislocation cores. The ferromagnetic dislocations have high coercivity due to their strong interaction with the surrounding antiferromagnetic bulk phase. Although it has already been reported that nanocrystals of rock-salt NiO show ferromagnetic behaviour, our study characterizes the ferromagnetic properties of individual lattice defects. We discuss the origin of the unexpected ferromagnetism in terms of the physical properties of the atomic-scale core structures of single dislocations, and demonstrate that it is possible to fabricate stable nanoscale magnetic elements inside crystalline environments composed of these microstructures.

No abstract

No abstract