arXiv subjects

Wenye Li

Publications and source records attributed to Wenye Li.

23 records · Page 2Linked to original sources

arXiv2020

Modeling Winner-Take-All Competition in Sparse Binary Projections

Inspired by the advances in biological science, the study of sparse binary projection models has attracted considerable recent research attention. The models project dense input samples into a higher-dimensional space and output sparse binary data representations after the Winner-Take-All competition, subject to the constraint that the projection matrix is also sparse and binary. Following the work along this line, we developed a supervised-WTA model when training samples with both input and output representations are available, from which the optimal projection matrix can be obtained with a simple, effective yet efficient algorithm. We further extended the model and the algorithm to an unsupervised setting where only the input representation of the samples is available. In a series of empirical evaluation on similarity search tasks, the proposed models reported significantly improved results over the state-of-the-art methods in both search accuracies and running speed. The successful results give us strong confidence that the work provides a highly practical tool to real world applications.

cs.LG↗

arXiv2019

Sparse Lifting of Dense Vectors: Unifying Word and Sentence Representations

As the first step in automated natural language processing, representing words and sentences is of central importance and has attracted significant research attention. Different approaches, from the early one-hot and bag-of-words representation to more recent distributional dense and sparse representations, were proposed. Despite the successful results that have been achieved, such vectors tend to consist of uninterpretable components and face nontrivial challenge in both memory and computational requirement in practical applications. In this paper, we designed a novel representation model that projects dense word vectors into a higher dimensional space and favors a highly sparse and binary representation of word vectors with potentially interpretable components, while trying to maintain pairwise inner products between original vectors as much as possible. Computationally, our model is relaxed as a symmetric non-negative matrix factorization problem which admits a fast yet effective solution. In a series of empirical evaluations, the proposed model exhibited consistent improvement and high potential in practical applications.

cs.CL↗

arXiv2019

An Augmented Transformer Architecture for Natural Language Generation Tasks

The Transformer based neural networks have been showing significant advantages on most evaluations of various natural language processing and other sequence-to-sequence tasks due to its inherent architecture based superiorities. Although the main architecture of the Transformer has been continuously being explored, little attention was paid to the positional encoding module. In this paper, we enhance the sinusoidal positional encoding algorithm by maximizing the variances between encoded consecutive positions to obtain additional promotion. Furthermore, we propose an augmented Transformer architecture encoded with additional linguistic knowledge, such as the Part-of-Speech (POS) tagging, to boost the performance on some natural language generation tasks, e.g., the automatic translation and summarization tasks. Experiments show that the proposed architecture attains constantly superior results compared to the vanilla Transformer.

cs.CL↗

arXiv2019

AuxBlocks: Defense Adversarial Example via Auxiliary Blocks

Deep learning models are vulnerable to adversarial examples, which poses an indisputable threat to their applications. However, recent studies observe gradient-masking defenses are self-deceiving methods if an attacker can realize this defense. In this paper, we propose a new defense method based on appending information. We introduce the Aux Block model to produce extra outputs as a self-ensemble algorithm and analytically investigate the robustness mechanism of Aux Block. We have empirically studied the efficiency of our method against adversarial examples in two types of white-box attacks, and found that even in the full white-box attack where an adversary can craft malicious examples from defense models, our method has a more robust performance of about 54.6% precision on Cifar10 dataset and 38.7% precision on Mini-Imagenet dataset. Another advantage of our method is that it is able to maintain the prediction accuracy of the classification model on clean images, and thereby exhibits its high potential in practical applications

cs.LG↗

arXiv2018

Network Community Detection with A Successive Spectral Relaxation Method

With invaluable theoretical and practical benefits, the problem of partitioning networks for community structures has attracted significant research attention in scientific and engineering disciplines. In literature, Newman's modularity measure is routinely applied to quantify the quality of a given partition, and thereby maximizing the measure provides a principled way of detecting communities in networks. Unfortunately, the exact optimization of the measure is computationally NP-complete and only applicable to very small networks. Approximation approaches have to be sought to scale to large networks. To address the computational issue, we proposed a new method to identify the partition decisions. Coupled with an iterative rounding strategy and a fast constrained power method, our work achieves tight and effective spectral relaxations. The proposed method was evaluated thoroughly on both real and synthetic networks. Compared with state-of-the-art approaches, the method obtained comparable, if not better, qualities. Meanwhile, it is highly suitable for parallel execution and reported a nearly linear improvement in running speed when increasing the number of computing nodes, which thereby provides a practical tool for partitioning very large networks.

cs.SI↗