It has been recently shown that a convolutional neural network can learn optical flow estimation with unsupervised learning. However, the performance of the unsupervised methods still has a relatively large gap compared to its supervised counterpart. Occlusion and large motion are some of the major factors that limit the current unsupervised learning of optical flow methods. In this work we introduce a new method which models occlusion explicitly and a new warping way that facilitates the learning of large motion. Our method shows promising results on Flying Chairs, MPI-Sintel and KITTI benchmark datasets. Especially on KITTI dataset where abundant unlabeled samples exist, our unsupervised method outperforms its counterpart trained with supervised learning.

Compositional generalization is a basic mechanism in human language learning, but current neural networks lack such ability. In this paper, we conduct fundamental research for encoding compositionality in neural networks. Conventional methods use a single representation for the input sentence, making it hard to apply prior knowledge of compositionality. In contrast, our approach leverages such knowledge with two representations, one generating attention maps, and the other mapping attended input words to output symbols. We reduce the entropy in each representation to improve generalization. Our experiments demonstrate significant improvements over the conventional methods in five NLP tasks including instruction learning and machine translation. In the SCAN domain, it boosts accuracies from 14.0% to 98.8% in Jump task, and from 92.0% to 99.7% in TurnLeft task. It also beats human performance on a few-shot learning task. We hope the proposed approach can help ease future research towards human-level compositional language learning. Source code is available online 1 .

No abstract

No abstract

As important regulators of cellular signal transduction, members of the protein tyrosine phosphatase (PTP) family are considered to be promising drug targets. However, to date, the most effective in vitro PTP inhibitors have tended to be highly charged, thus limiting cellular permeability. Here, we have identified an uncharged thioxothiazolidinone derivative (compound 1), as a competitive inhibitor of a subset of PTPs. Compound 1 effectively inhibited protein tyrosine phosphatase 1B (PTP1B) in two cell-based systems: it sensitized wild-type, but not PTP1B-null fibroblasts to insulin stimulation and prevented PTP1B-dependent dephosphorylation of the FLT3-ITD receptor tyrosine kinase. We have also tested a series of derivatives in vitro against PTP1B and proposed a model of the PTP1B-inhibitor interaction. These compounds should be useful in the elucidation of cellular PTP function and could represent a starting point for development of therapeutic PTP inhibitors.