Labeling training data has become one of the major roadblocks to using machine learning. Among various weak supervision paradigms, programmatic weak supervision (PWS) has achieved remarkable success in easing the manual labeling bottleneck by programmatically synthesizing training labels from multiple potentially noisy supervision sources. This paper presents a comprehensive survey of recent advances in PWS. In particular, we give a brief introduction of the PWS learning paradigm, and review representative approaches for each component within PWS's learning workflow. In addition, we discuss complementary learning paradigms for tackling limited labeled data scenarios and how these related approaches can be used in conjunction with PWS. Finally, we identify several critical challenges that remain under-explored in the area to hopefully inspire future research directions in the field.
Graph matching finds the correspondence of nodes across two correlated graphs and lies at the core of many applications. When graph side information is not available, the node correspondence is estimated on the sole basis of network topologies. In this paper, we propose a novel criterion to measure the graph matching accuracy, structural inconsistency (SI), which is defined based on the network topological structure. Specifically, SI incorporates the heat diffusion wavelet to accommodate the multi-hop structure of the graphs. Based on SI, we propose a Structural Inconsistency reducing Graph Matching Algorithm (SIGMA), which improves the alignment scores of node pairs that have low SI values in each iteration. Under suitable assumptions, SIGMA can reduce SI values of true counterparts. Furthermore, we demonstrate that SIGMA can be derived by using a mirror descent method to solve the Gromov-Wasserstein distance with a novel K-hop-structure-based matching costs. Extensive experiments show that our method outperforms state-of-the-art methods.
The recurrent neural network transducer (RNN-T) has recently become the mainstream end-to-end approach for streaming automatic speech recognition (ASR). To estimate the output distributions over subword units, RNN-T uses a fully connected layer as the joint network to fuse the acoustic representations extracted using the acoustic encoder with the text representations obtained using the prediction network based on the previous subword units. In this paper, we propose to use gating, bilinear pooling, and a combination of them in the joint network to produce more expressive representations to feed into the output layer. A regularisation method is also proposed to enable better acoustic encoder training by reducing the gradients back-propagated into the prediction network at the beginning of RNN-T training. Experimental results on a multilingual ASR setting for voice search over nine languages show that the joint use of the proposed methods can result in 4%--5% relative word error rate reductions with only a few million extra parameters.
Transfer-based adversarial example is one of the most important classes of black-box attacks. However, there is a trade-off between transferability and imperceptibility of the adversarial perturbation. Prior work in this direction often requires a fixed but large $\ell_p$-norm perturbation budget to reach a good transfer success rate, leading to perceptible adversarial perturbations. On the other hand, most of the current unrestricted adversarial attacks that aim to generate semantic-preserving perturbations suffer from weaker transferability to the target model. In this work, we propose a geometry-aware framework to generate transferable adversarial examples with minimum changes. Analogous to model selection in statistical machine learning, we leverage a validation model to select the optimal perturbation budget for each image under both the $\ell_{\infty}$-norm and unrestricted threat models. Extensive experiments verify the effectiveness of our framework on balancing imperceptibility and transferability of the crafted adversarial examples. The methodology is the foundation of our entry to the CVPR'21 Security AI Challenger: Unrestricted Adversarial Attacks on ImageNet, in which we ranked 1st place out of 1,559 teams and surpassed the runner-up submissions by 4.59% and 23.91% in terms of final score and average image quality level, respectively. Code is available at https://github.com/Equationliu/GA-Attack.
Modern high-dimensional methods often adopt the ``bet on sparsity'' principle, while in supervised multivariate learning statisticians may face ``dense'' problems with a large number of nonzero coefficients. This paper proposes a novel clustered reduced-rank learning (CRL) framework that imposes two joint matrix regularizations to automatically group the features in constructing predictive factors. CRL is more interpretable than low-rank modeling and relaxes the stringent sparsity assumption in variable selection. In this paper, new information-theoretical limits are presented to reveal the intrinsic cost of seeking for clusters, as well as the blessing from dimensionality in multivariate learning. Moreover, an efficient optimization algorithm is developed, which performs subspace learning and clustering with guaranteed convergence. The obtained fixed-point estimators, though not necessarily globally optimal, enjoy the desired statistical accuracy beyond the standard likelihood setup under some regularity conditions. Moreover, a new kind of information criterion, as well as its scale-free form, is proposed for cluster and rank selection, and has a rigorous theoretical support without assuming an infinite sample size. Extensive simulations and real-data experiments demonstrate the statistical accuracy and interpretability of the proposed method.
Despite the great success of pre-trained language models (LMs) in many natural language processing (NLP) tasks, they require excessive labeled data for fine-tuning to achieve satisfactory performance. To enhance the label efficiency, researchers have resorted to active learning (AL), while the potential of unlabeled data is ignored by most of prior work. To unleash the power of unlabeled data for better label efficiency and model performance, we develop ATM, a new framework that leverage self-training to exploit unlabeled data and is agnostic to the specific AL algorithm, serving as a plug-in module to improve existing AL methods. Specifically, the unlabeled data with high uncertainty is exposed to oracle for annotations while those with low uncertainty are leveraged for self-training. To alleviate the label noise propagation issue in self-training, we design a simple and effective momentum-based memory bank to dynamically aggregate the model predictions from all rounds. By extensive experiments, we demonstrate that ATM outperforms the strongest active learning and self-training baselines and improve the label efficiency by 51.9% on average.
The recently developed Particle-based Variational Inference (ParVI) methods drive the empirical distribution of a set of \emph{fixed-weight} particles towards a given target distribution $\pi$ by iteratively updating particles' positions. However, the fixed weight restriction greatly confines the empirical distribution's approximation ability, especially when the particle number is limited. In this paper, we propose to dynamically adjust particles' weights according to a Fisher-Rao reaction flow. We develop a general Dynamic-weight Particle-based Variational Inference (DPVI) framework according to a novel continuous composite flow, which evolves the positions and weights of particles simultaneously. We show that the mean-field limit of our composite flow is actually a Wasserstein-Fisher-Rao gradient flow of certain dissimilarity functional $\mathcal{F}$, which leads to a faster decrease of $\mathcal{F}$ than the Wasserstein gradient flow underlying existing fixed-weight ParVIs. By using different finite-particle approximations in our general framework, we derive several efficient DPVI algorithms. The empirical results demonstrate the superiority of our derived DPVI algorithms over their fixed-weight counterparts.
Subspace clustering methods embrace a self-expressive model that represents each data point as a linear combination of other data points in the dataset are powerful unsupervised learning techniques. However, when dealing with large-scale datasets, the representation of each data point by referring to all data points as a dictionary suffers from high computational complexity. To alleviate this issue, we introduce a parallelizable multi-subset based self-expressive model (PMS) which represents each data point by combing multiple subsets, with each consisting of only a small percentage of samples. The adoption of PMS in subspace clustering (PMSSC) leads to computational advantages because each optimization problem decomposed into each subset is small, and can be solved efficiently in parallel. Besides, PMSSC is able to combine multiple self-expressive coefficient vectors obtained from subsets, which contributes to the improvement of self-expressiveness. Extensive experiments on synthetic data and real-world datasets show the efficiency and effectiveness of our approach against competitive methods.
The growing availability of the data collected from smart manufacturing is changing the paradigms of production monitoring and control. The increasing complexity and content of the wafer manufacturing process in addition to the time-varying unexpected disturbances and uncertainties, make it infeasible to do the control process with model-based approaches. As a result, data-driven soft-sensing modeling has become more prevalent in wafer process diagnostics. Recently, deep learning has been utilized in soft sensing system with promising performance on highly nonlinear and dynamic time-series data. Despite its successes in soft-sensing systems, however, the underlying logic of the deep learning framework is hard to understand. In this paper, we propose a deep learning-based model for defective wafer detection using a highly imbalanced dataset. To understand how the proposed model works, the deep visualization approach is applied. Additionally, the model is then fine-tuned guided by the deep visualization. Extensive experiments are performed to validate the effectiveness of the proposed system. The results provide an interpretation of how the model works and an instructive fine-tuning method based on the interpretation.
Over the last few decades, modern industrial processes have investigated several cost-effective methodologies to improve the productivity and yield of semiconductor manufacturing. While playing an essential role in facilitating real-time monitoring and control, the data-driven soft-sensors in industries have provided a competitive edge when augmented with deep learning approaches for wafer fault-diagnostics. Despite the success of deep learning methods across various domains, they tend to suffer from bad performance on multi-variate soft-sensing data domains. To mitigate this, we propose a soft-sensing ConFormer (CONvolutional transFORMER) for wafer fault-diagnostic classification task which primarily consists of multi-head convolution modules that reap the benefits of fast and light-weight operations of convolutions, and also the ability to learn the robust representations through multi-head design alike transformers. Another key issue is that traditional learning paradigms tend to suffer from low performance on noisy and highly-imbalanced soft-sensing data. To address this, we augment our soft-sensing ConFormer model with a curriculum learning-based loss function, which effectively learns easy samples in the early phase of training and difficult ones later. To further demonstrate the utility of our proposed architecture, we performed extensive experiments on various toolsets of Seagate Technology's wafer manufacturing process which are shared openly along with this work. To the best of our knowledge, this is the first time that curriculum learning-based soft-sensing ConFormer architecture has been proposed for soft-sensing data and our results show strong promise for future use in soft-sensing research domain.