arXiv subjects

Ke Li

Publications and source records attributed to Ke Li.

At least 379 records · Page 21Linked to original sources

arXiv2021

A Parallelizable Lattice Rescoring Strategy with Neural Language Models

This paper proposes a parallel computation strategy and a posterior-based lattice expansion algorithm for efficient lattice rescoring with neural language models (LMs) for automatic speech recognition. First, lattices from first-pass decoding are expanded by the proposed posterior-based lattice expansion algorithm. Second, each expanded lattice is converted into a minimal list of hypotheses that covers every arc. Each hypothesis is constrained to be the best path for at least one arc it includes. For each lattice, the neural LM scores of the minimal list are computed in parallel and are then integrated back to the lattice in the rescoring stage. Experiments on the Switchboard dataset show that the proposed rescoring strategy obtains comparable recognition performance and generates more compact lattices than a competitive baseline method. Furthermore, the parallel rescoring method offers more flexibility by simplifying the integration of PyTorch-trained neural LMs for lattice rescoring with Kaldi.

eess.AS↗

arXiv2021

Filter Grafting for Deep Neural Networks: Reason, Method, and Cultivation

Filter is the key component in modern convolutional neural networks (CNNs). However, since CNNs are usually over-parameterized, a pre-trained network always contain some invalid (unimportant) filters. These filters have relatively small $l_{1}$ norm and contribute little to the output (\textbf{Reason}). While filter pruning removes these invalid filters for efficiency consideration, we tend to reactivate them to improve the representation capability of CNNs. In this paper, we introduce filter grafting (\textbf{Method}) to achieve this goal. The activation is processed by grafting external information (weights) into invalid filters. To better perform the grafting, we develop a novel criterion to measure the information of filters and an adaptive weighting strategy to balance the grafted information among networks. After the grafting operation, the network has fewer invalid filters compared with its initial state, enpowering the model with more representation capacity. Meanwhile, since grafting is operated reciprocally on all networks involved, we find that grafting may lose the information of valid filters when improving invalid filters. To gain a universal improvement on both valid and invalid filters, we compensate grafting with distillation (\textbf{Cultivation}) to overcome the drawback of grafting . Extensive experiments are performed on the classification and recognition tasks to show the superiority of our method. Code is available at \textcolor{black}{\emph{https://github.com/fxmeng/filter-grafting}}.

cs.LG↗

arXiv2020

Enhancing Unsupervised Video Representation Learning by Decoupling the Scene and the Motion

One significant factor we expect the video representation learning to capture, especially in contrast with the image representation learning, is the object motion. However, we found that in the current mainstream video datasets, some action categories are highly related with the scene where the action happens, making the model tend to degrade to a solution where only the scene information is encoded. For example, a trained model may predict a video as playing football simply because it sees the field, neglecting that the subject is dancing as a cheerleader on the field. This is against our original intention towards the video representation learning and may bring scene bias on different dataset that can not be ignored. In order to tackle this problem, we propose to decouple the scene and the motion (DSM) with two simple operations, so that the model attention towards the motion information is better paid. Specifically, we construct a positive clip and a negative clip for each video. Compared to the original video, the positive/negative is motion-untouched/broken but scene-broken/untouched by Spatial Local Disturbance and Temporal Local Disturbance. Our objective is to pull the positive closer while pushing the negative farther to the original clip in the latent space. In this way, the impact of the scene is weakened while the temporal sensitivity of the network is further enhanced. We conduct experiments on two tasks with various backbones and different pre-training datasets, and find that our method surpass the SOTA methods with a remarkable 8.1% and 8.8% improvement towards action recognition task on the UCF101 and HMDB51 datasets respectively using the same backbone.

cs.CV↗

arXiv2020

One for More: Selecting Generalizable Samples for Generalizable ReID Model

Current training objectives of existing person Re-IDentification (ReID) models only ensure that the loss of the model decreases on selected training batch, with no regards to the performance on samples outside the batch. It will inevitably cause the model to over-fit the data in the dominant position (e.g., head data in imbalanced class, easy samples or noisy samples). %We call the sample that updates the model towards generalizing on more data a generalizable sample. The latest resampling methods address the issue by designing specific criterion to select specific samples that trains the model generalize more on certain type of data (e.g., hard samples, tail data), which is not adaptive to the inconsistent real world ReID data distributions. Therefore, instead of simply presuming on what samples are generalizable, this paper proposes a one-for-more training objective that directly takes the generalization ability of selected samples as a loss function and learn a sampler to automatically select generalizable samples. More importantly, our proposed one-for-more based sampler can be seamlessly integrated into the ReID training framework which is able to simultaneously train ReID models and the sampler in an end-to-end fashion. The experimental results show that our method can effectively improve the ReID model training and boost the performance of ReID models.

cs.CV↗

arXiv2020

Pruning Filter in Filter

Pruning has become a very powerful and effective technique to compress and accelerate modern neural networks. Existing pruning methods can be grouped into two categories: filter pruning (FP) and weight pruning (WP). FP wins at hardware compatibility but loses at the compression ratio compared with WP. To converge the strength of both methods, we propose to prune the filter in the filter. Specifically, we treat a filter $F \in \mathbb{R}^{C\times K\times K}$ as $K \times K$ stripes, i.e., $1\times 1$ filters $\in \mathbb{R}^{C}$, then by pruning the stripes instead of the whole filter, we can achieve finer granularity than traditional FP while being hardware friendly. We term our method as SWP (\emph{Stripe-Wise Pruning}). SWP is implemented by introducing a novel learnable matrix called Filter Skeleton, whose values reflect the shape of each filter. As some recent work has shown that the pruned architecture is more crucial than the inherited important weights, we argue that the architecture of a single filter, i.e., the shape, also matters. Through extensive experiments, we demonstrate that SWP is more effective compared to the previous FP-based methods and achieves the state-of-art pruning ratio on CIFAR-10 and ImageNet datasets without obvious accuracy drop. Code is available at https://github.com/fxmeng/Pruning-Filter-in-Filter

cs.CV↗

arXiv2020

Generating Unobserved Alternatives

We consider problems where multiple predictions can be considered correct, but only one of them is given as supervision. This setting differs from both the regression and class-conditional generative modelling settings: in the former, there is a unique observed output for each input, which is provided as supervision; in the latter, there are many observed outputs for each input, and many are provided as supervision. Applying either regression methods and conditional generative models to the present setting often results in a model that can only make a single prediction for each input. We explore several problems that have this property and develop an approach that can generate multiple high-quality predictions given the same input. As a result, it can be used to generate high-quality outputs that are different from the observed output.

cs.LG↗

arXiv2020

Better Knowledge Retention through Metric Learning

In continual learning, new categories may be introduced over time, and an ideal learning system should perform well on both the original categories and the new categories. While deep neural nets have achieved resounding success in the classical supervised setting, they are known to forget about knowledge acquired in prior episodes of learning if the examples encountered in the current episode of learning are drastically different from those encountered in prior episodes. In this paper, we propose a new method that can both leverage the expressive power of deep neural nets and is resilient to forgetting when new categories are introduced. We found the proposed method can reduce forgetting by 2.3x to 6.9x on CIFAR-10 compared to existing methods and by 1.8x to 2.7x on ImageNet compared to an oracle baseline.

cs.LG↗

arXiv2020

DeRF: Decomposed Radiance Fields

With the advent of Neural Radiance Fields (NeRF), neural networks can now render novel views of a 3D scene with quality that fools the human eye. Yet, generating these images is very computationally intensive, limiting their applicability in practical scenarios. In this paper, we propose a technique based on spatial decomposition capable of mitigating this issue. Our key observation is that there are diminishing returns in employing larger (deeper and/or wider) networks. Hence, we propose to spatially decompose a scene and dedicate smaller networks for each decomposed part. When working together, these networks can render the whole scene. This allows us near-constant inference time regardless of the number of decomposed parts. Moreover, we show that a Voronoi spatial decomposition is preferable for this purpose, as it is provably compatible with the Painter's Algorithm for efficient and GPU-friendly rendering. Our experiments show that for real-world scenes, our method provides up to 3x more efficient inference than NeRF (with the same rendering quality), or an improvement of up to 1.0~dB in PSNR (for the same inference cost).

cs.CV↗

arXiv2020

University of Washington at TREC 2020 Fairness Ranking Track

InfoSeeking Lab's FATE (Fairness Accountability Transparency Ethics) group at University of Washington participated in 2020 TREC Fairness Ranking Track. This report describes that track, assigned data and tasks, our group definitions, and our results. Our approach to bringing fairness in retrieval and re-ranking tasks with Semantic Scholar data was to extract various dimensions of author identity. These dimensions included gender and location. We developed modules for these extractions in a way that allowed us to plug them in for either of the tasks as needed. After trying different combinations of relative weights assigned to relevance, gender, and location information, we chose five runs for retrieval and five runs for re-ranking tasks. The results showed that our runs performed below par for re-ranking task, but above average for retrieval.

cs.IR↗

arXiv2020

A Multi-stream Convolutional Neural Network for Micro-expression Recognition Using Optical Flow and EVM

Micro-expression (ME) recognition plays a crucial role in a wide range of applications, particularly in public security and psychotherapy. Recently, traditional methods rely excessively on machine learning design and the recognition rate is not high enough for its practical application because of its short duration and low intensity. On the other hand, some methods based on deep learning also cannot get high accuracy due to problems such as the imbalance of databases. To address these problems, we design a multi-stream convolutional neural network (MSCNN) for ME recognition in this paper. Specifically, we employ EVM and optical flow to magnify and visualize subtle movement changes in MEs and extract the masks from the optical flow images. And then, we add the masks, optical flow images, and grayscale images into the MSCNN. After that, in order to overcome the imbalance of databases, we added a random over-sampler after the Dense Layer of the neural network. Finally, extensive experiments are conducted on two public ME databases: CASME II and SAMM. Compared with many recent state-of-the-art approaches, our method achieves more promising recognition results.

cs.CV↗

arXiv2020

Neural Language Modeling With Implicit Cache Pointers

A cache-inspired approach is proposed for neural language models (LMs) to improve long-range dependency and better predict rare words from long contexts. This approach is a simpler alternative to attention-based pointer mechanism that enables neural LMs to reproduce words from recent history. Without using attention and mixture structure, the method only involves appending extra tokens that represent words in history to the output layer of a neural LM and modifying training supervisions accordingly. A memory-augmentation unit is introduced to learn words that are particularly likely to repeat. We experiment with both recurrent neural network- and Transformer-based LMs. Perplexity evaluation on Penn Treebank and WikiText-2 shows the proposed model outperforms both LSTM and LSTM with attention-based pointer mechanism and is more effective on rare words. N-best rescoring experiments on Switchboard indicate that it benefits both very rare and frequent words. However, it is challenging for the proposed model as well as two other models with attention-based pointer mechanism to obtain good overall WER reductions.

eess.AS↗

arXiv2020

BiLO-CPDP: Bi-Level Programming for Automated Model Discovery in Cross-Project Defect Prediction

Cross-Project Defect Prediction (CPDP), which borrows data from similar projects by combining a transfer learner with a classifier, have emerged as a promising way to predict software defects when the available data about the target project is insufficient. How-ever, developing such a model is challenge because it is difficult to determine the right combination of transfer learner and classifier along with their optimal hyper-parameter settings. In this paper, we propose a tool, dubbedBiLO-CPDP, which is the first of its kind to formulate the automated CPDP model discovery from the perspective of bi-level programming. In particular, the bi-level programming proceeds the optimization with two nested levels in a hierarchical manner. Specifically, the upper-level optimization routine is designed to search for the right combination of transfer learner and classifier while the nested lower-level optimization routine aims to optimize the corresponding hyper-parameter settings.To evaluateBiLO-CPDP, we conduct experiments on 20 projects to compare it with a total of 21 existing CPDP techniques, along with its single-level optimization variant and Auto-Sklearn, a state-of-the-art automated machine learning tool. Empirical results show that BiLO-CPDP champions better prediction performance than all other 21 existing CPDP techniques on 70% of the projects, while be-ing overwhelmingly superior to Auto-Sklearn and its single-level optimization variant on all cases. Furthermore, the unique bi-level formalization inBiLO-CPDP also permits to allocate more budget to the upper-level, which significantly boosts the performance.

cs.SE↗

arXiv2020

Inclusive GAN: Improving Data and Minority Coverage in Generative Models

Generative Adversarial Networks (GANs) have brought about rapid progress towards generating photorealistic images. Yet the equitable allocation of their modeling capacity among subgroups has received less attention, which could lead to potential biases against underrepresented minorities if left uncontrolled. In this work, we first formalize the problem of minority inclusion as one of data coverage, and then propose to improve data coverage by harmonizing adversarial training with reconstructive generation. The experiments show that our method outperforms the existing state-of-the-art methods in terms of data coverage on both seen and unseen data. We develop an extension that allows explicit control over the minority subgroups that the model should ensure to include, and validate its effectiveness at little compromise from the overall performance on the entire dataset. Code, models, and supplemental videos are available at GitHub.

cs.CV↗

arXiv2020

Search-Based Software Engineering for Self-Adaptive Systems: Survey, Disappointments, Suggestions and Opportunities

Search-Based Software Engineering (SBSE) is a promising paradigm that exploits the computational search to optimize different processes when engineering complex software systems. Self-adaptive system (SAS) is one category of such complex systems that permits to optimize different functional and non-functional objectives/criteria under changing environments (e.g., requirements and workload), which involves problems that are subject to search. In this regard, over years, there has been a considerable amount of work that investigates SBSE for SASs. In this paper, we provide the first systematic and comprehensive survey exclusively on SBSE for SASs, covering papers in 27 venues from 7 repositories, which eventually leads to several key statistics from the most notable 74 primary studies in this particular field of research. Our results, surprisingly, have revealed five disappointments that are of utmost importance and can result in serve consequences but have been overwhelmingly ignored in existing studies. We provide theoretical and/or experimental evidence to justify our arguments against the disappointments, present suggestions, and highlight the promising research opportunities towards their mitigation. We also elaborate on three other emergent, but currently under-explored opportunities for future work on SBSE for SASs. By mitigating the disappointments revealed in this work, together with the highlighted opportunities, we hope to be able to excite a much more significant growth in this particular research direction.

cs.SE↗

arXiv2020

Efficient MDI Adaptation for n-gram Language Models

This paper presents an efficient algorithm for n-gram language model adaptation under the minimum discrimination information (MDI) principle, where an out-of-domain language model is adapted to satisfy the constraints of marginal probabilities of the in-domain data. The challenge for MDI language model adaptation is its computational complexity. By taking advantage of the backoff structure of n-gram model and the idea of hierarchical training method, originally proposed for maximum entropy (ME) language models, we show that MDI adaptation can be computed in linear-time complexity to the inputs in each iteration. The complexity remains the same as ME models, although MDI is more general than ME. This makes MDI adaptation practical for large corpus and vocabulary. Experimental results confirm the scalability of our algorithm on very large datasets, while MDI adaptation gets slightly worse perplexity but better word error rate results compared to simple linear interpolation.

cs.CL↗

arXiv2020

Highly Efficient Second Harmonic Generation of Thin Film Lithium Niobate Nanograting near Bound States in the Continuum

Bound states in the continuum (BICs), a concept from quantum mechanics, are ubiquitous physical phenomena where waves will be completely locked inside physical systems without energy leaky. Such a physical phenomenon in optics will provide a platform for optical mode confinement to strengthen local field enhancement in nonlinear optics. Here we utilize an optical system consisting of asymmetric nanogratings and waveguide of thin film lithium niobate (LiNbO3) material to enhance second harmonic response near BICs. By breaking the symmetry of grating periodicity, we realize strong local field confined inside waveguide up to 25 times normalized to incident field (with dissymmetric factor of 0.2), allowing strong light-matter interaction in nonlinear material. From the numerical simulation, we theoretically demonstrate that such an optical system can greatly enhance second harmonic intensity enhancement of about 104 compared with undersigned LiNbO3 film and conversion efficiency reaching 1.53e-5 for dissymmetric factor=0.2 under illumination of 1.33 GW/(suqare cm). Surprisingly, we can predict that a giant enhancement of second harmonic conversion efficiency will exceed 8.13e-5 for dissymmetric factor=0.1 when the optical system is extremely close to BICs. We believe that such an optical system to trap local field inside is also accessible to promote the application of thin film lithium niobate in the field of integrated nonlinear optics.

physics.optics↗

arXiv2020

The JHU Multi-Microphone Multi-Speaker ASR System for the CHiME-6 Challenge

This paper summarizes the JHU team's efforts in tracks 1 and 2 of the CHiME-6 challenge for distant multi-microphone conversational speech diarization and recognition in everyday home environments. We explore multi-array processing techniques at each stage of the pipeline, such as multi-array guided source separation (GSS) for enhancement and acoustic model training data, posterior fusion for speech activity detection, PLDA score fusion for diarization, and lattice combination for automatic speech recognition (ASR). We also report results with different acoustic model architectures, and integrate other techniques such as online multi-channel weighted prediction error (WPE) dereverberation and variational Bayes-hidden Markov model (VB-HMM) based overlap assignment to deal with reverberation and overlapping speakers, respectively. As a result of these efforts, our ASR systems achieve a word error rate of 40.5% and 67.5% on tracks 1 and 2, respectively, on the evaluation set. This is an improvement of 10.8% and 10.4% absolute, over the challenge baselines for the respective tracks.

eess.AS↗

arXiv2020

DeepSQLi: Deep Semantic Learning for Testing SQL Injection

Security is unarguably the most serious concern for Web applications, to which SQL injection (SQLi) attack is one of the most devastating attacks. Automatically testing SQLi vulnerabilities is of ultimate importance, yet is unfortunately far from trivial to implement. This is because the existence of a huge, or potentially infinite, number of variants and semantic possibilities of SQL leading to SQLi attacks on various Web applications. In this paper, we propose a deep natural language processing based tool, dubbed DeepSQLi, to generate test cases for detecting SQLi vulnerabilities. Through adopting deep learning based neural language model and sequence of words prediction, DeepSQLi is equipped with the ability to learn the semantic knowledge embedded in SQLi attacks, allowing it to translate user inputs (or a test case) into a new test case, which is semantically related and potentially more sophisticated. Experiments are conducted to compare DeepSQLi with SQLmap, a state-of-the-art SQLi testing automation tool, on six real-world Web applications that are of different scales, characteristics and domains. Empirical results demonstrate the effectiveness and the remarkable superiority of DeepSQLi over SQLmap, such that more SQLi vulnerabilities can be identified by using a less number of test cases, whilst running much faster.

cs.SE↗