Aspect-based sentiment analysis (ABSA) is a fine-grained sentiment analysis task which aims to extract the aspects from sentences and identify their corresponding sentiments. Aspect term extraction (ATE) is the crucial step for ABSA. Due to the expensive annotation for aspect terms, we often lack labeled target domain data for fine-tuning. To address this problem, many approaches have been proposed recently to transfer common knowledge in an unsupervised way, but such methods have too many modules and require expensive multi-stage preprocessing. In this paper, we propose a simple but effective technique based on mutual information maximization, which can serve as an additional component to enhance any kind of model for cross-domain ABSA and ATE. Furthermore, we provide some analysis of this approach. Experiment results show that our proposed method outperforms the state-of-the-art methods for cross-domain ABSA by 4.32% Micro-F1 on average over 10 different domain pairs. Apart from that, our method can be extended to other sequence labeling tasks, such as named entity recognition (NER).
The structure and dynamics of isolated nanosamples in free flight can be directly visualized via single-shot coherent diffractive imaging using the intense and short pulses of X-ray free-electron lasers. Wide-angle scattering images even encode three-dimensional morphological information of the samples, but the retrieval of this information remains a challenge. Up to now, effective three-dimensional morphology reconstructions from single shots were only achieved via fitting with highly constrained models, requiring a priori knowledge about possible geometrical shapes. Here we present a much more generic imaging approach. Relying on a model that allows for any sample morphology described by a convex polyhedron, we reconstruct wide-angle diffraction patterns from individual silver nanoparticles. In addition to known structural motives with high symmetries, we retrieve imperfect shapes and agglomerates which were not accessible previously. Our results open new routes towards true 3D structure determination of single nanoparticles and, ultimately, 3D movies of ultrafast nanoscale dynamics.
We consider the multi-user detection (MUD) problem in uplink grant-free non-orthogonal multiple access (NOMA), where the access point has to identify the total number and correct identity of the active Internet of Things (IoT) devices and decode their transmitted data. We assume that IoT devices use complex spreading sequences and transmit information in a random-access manner following the burst-sparsity model, where some IoT devices transmit their data in multiple adjacent time slots with a high probability, while others transmit only once during a frame. Exploiting the temporal correlation, we propose an attention-based bidirectional long short-term memory (BiLSTM) network to solve the MUD problem. The BiLSTM network creates a pattern of the device activation history using forward and reverse pass LSTMs, whereas the attention mechanism provides essential context to the device activation points. By doing so, a hierarchical pathway is followed for detecting active devices in a grant-free scenario. Then, by utilising the complex spreading sequences, blind data detection for the estimated active devices is performed. The proposed framework does not require prior knowledge of device sparsity levels and channels for performing MUD. The results show that the proposed network achieves better performance compared to existing benchmark schemes.
Cross-entropy loss with softmax output is a standard choice to train neural network classifiers. We give a new view of neural network classifiers with softmax and cross-entropy as mutual information evaluators. We show that when the dataset is balanced, training a neural network with cross-entropy maximises the mutual information between inputs and labels through a variational form of mutual information. Thereby, we develop a new form of softmax that also converts a classifier to a mutual information evaluator when the dataset is imbalanced. Experimental results show that the new form leads to better classification accuracy, in particular for imbalanced datasets.
Recent studies have shown that detectors based on deep models are vulnerable to adversarial examples, even in the black-box scenario where the attacker cannot access the model information. Most existing attack methods aim to minimize the true positive rate, which often shows poor attack performance, as another sub-optimal bounding box may be detected around the attacked bounding box to be the new true positive one. To settle this challenge, we propose to minimize the true positive rate and maximize the false positive rate, which can encourage more false positive objects to block the generation of new true positive bounding boxes. It is modeled as a multi-objective optimization (MOP) problem, of which the generic algorithm can search the Pareto-optimal. However, our task has more than two million decision variables, leading to low searching efficiency. Thus, we extend the standard Genetic Algorithm with Random Subset selection and Divide-and-Conquer, called GARSDC, which significantly improves the efficiency. Moreover, to alleviate the sensitivity to population quality in generic algorithms, we generate a gradient-prior initial population, utilizing the transferability between different detectors with similar backbones. Compared with the state-of-art attack methods, GARSDC decreases by an average 12.0 in the mAP and queries by about 1000 times in extensive experiments. Our codes can be found at https://github.com/LiangSiyuan21/ GARSDC.
Intuition might suggest that motion and dynamic information are key to video-based action recognition. In contrast, there is evidence that state-of-the-art deep-learning video understanding architectures are biased toward static information available in single frames. Presently, a methodology and corresponding dataset to isolate the effects of dynamic information in video are missing. Their absence makes it difficult to understand how well contemporary architectures capitalize on dynamic vs. static information. We respond with a novel Appearance Free Dataset (AFD) for action recognition. AFD is devoid of static information relevant to action recognition in a single frame. Modeling of the dynamics is necessary for solving the task, as the action is only apparent through consideration of the temporal dimension. We evaluated 11 contemporary action recognition architectures on AFD as well as its related RGB video. Our results show a notable decrease in performance for all architectures on AFD compared to RGB. We also conducted a complimentary study with humans that shows their recognition accuracy on AFD and RGB is very similar and much better than the evaluated architectures on AFD. Our results motivate a novel architecture that revives explicit recovery of optical flow, within a contemporary design for best performance on AFD and RGB.
Lip reading aims to predict speech based on lip movements alone. As it focuses on visual information to model the speech, its performance is inherently sensitive to personal lip appearances and movements. This makes the lip reading models show degraded performance when they are applied to unseen speakers due to the mismatch between training and testing conditions. Speaker adaptation technique aims to reduce this mismatch between train and test speakers, thus guiding a trained model to focus on modeling the speech content without being intervened by the speaker variations. In contrast to the efforts made in audio-based speech recognition for decades, the speaker adaptation methods have not well been studied in lip reading. In this paper, to remedy the performance degradation of lip reading model on unseen speakers, we propose a speaker-adaptive lip reading method, namely user-dependent padding. The user-dependent padding is a speaker-specific input that can participate in the visual feature extraction stage of a pre-trained lip reading model. Therefore, the lip appearances and movements information of different speakers can be considered during the visual feature encoding, adaptively for individual speakers. Moreover, the proposed method does not need 1) any additional layers, 2) to modify the learned weights of the pre-trained model, and 3) the speaker label of train data used during pre-train. It can directly adapt to unseen speakers by learning the user-dependent padding only, in a supervised or unsupervised manner. Finally, to alleviate the speaker information insufficiency in public lip reading databases, we label the speaker of a well-known audio-visual database, LRW, and design an unseen-speaker lip reading scenario named LRW-ID.
In this paper, we propose SATformer, a novel Transformer-based solution for Boolean satisfiability (SAT) solving. Different from existing learning-based SAT solvers that learn at the problem instance level, SATformer learns the minimum unsatisfiable cores (MUC) of unsatisfiable problem instances, which provide rich information for the causality of such problems. Specifically, we apply a graph neural network (GNN) to obtain the embeddings of the clauses in the conjunctive normal format (CNF). A hierarchical Transformer architecture is applied on the clause embeddings to capture the relationships among clauses, and the self-attention weight is learned to be high when those clauses forming UNSAT cores are attended together, and set to be low otherwise. By doing so, SATformer effectively learns the correlations among clauses for SAT prediction. Experimental results show that SATformer is more powerful than existing end-to-end learning-based SAT solvers.
Recently, manipulation of real-world images has been highly elaborated along with the development of Generative Adversarial Networks (GANs) and corresponding encoders, which embed real-world images into the latent space. However, designing encoders of GAN still remains a challenging task due to the trade-off between distortion and perception. In this paper, we point out that the existing encoders try to lower the distortion not only on the interest region, e.g., human facial region but also on the uninterest region, e.g., background patterns and obstacles. However, most uninterest regions in real-world images are located at out-of-distribution (OOD), which are infeasible to be ideally reconstructed by generative models. Moreover, we empirically find that the uninterest region overlapped with the interest region can mangle the original feature of the interest region, e.g., a microphone overlapped with a facial region is inverted into the white beard. As a result, lowering the distortion of the whole image while maintaining the perceptual quality is very challenging. To overcome this trade-off, we propose a simple yet effective encoder training scheme, coined IntereStyle, which facilitates encoding by focusing on the interest region. IntereStyle steers the encoder to disentangle the encodings of the interest and uninterest regions. To this end, we filter the information of the uninterest region iteratively to regulate the negative impact of the uninterest region. We demonstrate that IntereStyle achieves both lower distortion and higher perceptual quality compared to the existing state-of-the-art encoders. Especially, our model robustly conserves features of the original images, which shows the robust image editing and style mixing results. We will release our code with the pre-trained model after the review.
In this paper, we introduce novel lightweight generative adversarial networks, which can effectively capture long-range dependencies in the image generation process, and produce high-quality results with a much simpler architecture. To achieve this, we first introduce a long-range module, allowing the network to dynamically adjust the number of focused sampling pixels and to also augment sampling locations. Thus, it can break the limitation of the fixed geometric structure of the convolution operator, and capture long-range dependencies in both spatial and channel-wise directions. Also, the proposed long-range module can highlight negative relations between pixels, working as a regularization to stabilize training. Furthermore, we propose a new generation strategy through which we introduce metadata into the image generation process to provide basic information about target images, which can stabilize and speed up the training process. Our novel long-range module only introduces few additional parameters and is easily inserted into existing models to capture long-range dependencies. Extensive experiments demonstrate the competitive performance of our method with a lightweight architecture.