In this paper, we study abstractive review summarization.Observing that review summaries often consist of aspect words, opinion words and context words, we propose a two-stage reinforcement learning approach, which first predicts the output word type from the three types, and then leverages the predicted word type to generate the final word distribution.Experimental results on two Amazon product review datasets demonstrate that our method can consistently outperform several strong baseline approaches based on ROUGE scores.
Recently, a few novel streaming attention-based sequence-to-sequence (S2S) models have been proposed to perform online speech recognition with linear-time decoding complexity. However, in these models, the decisions to generate tokens are delayed compared to the actual acoustic boundaries since their unidirectional encoders lack future information. This leads to an inevitable latency during inference. To alleviate this issue and reduce latency, we propose several strategies during training by leveraging external hard alignments extracted from the hybrid model. We investigate to utilize the alignments in both the encoder and the decoder. On the encoder side, (1) multi-task learning and (2) pre-training with the framewise classification task are studied. On the decoder side, we (3) remove inappropriate alignment paths beyond an acceptable latency during the alignment marginalization, and (4) directly minimize the differentiable expected latency loss. Experiments on the Cortana voice search task demonstrate that our proposed methods can significantly reduce the latency, and even improve the recognition accuracy in certain cases on the decoder side. We also present some analysis to understand the behaviors of streaming S2S models.
Our work expands the use of capsule networks to the task of object segmentation for the first time in the literature. This is made possible via the introduction of locally-constrained routing and transformation matrix sharing, which reduces the parameter/memory burden and allows for the segmentation of objects at large resolutions. To compensate for the loss of global information in constraining the routing, we propose the concept of "deconvolutional" capsules to create a deep encoder-decoder style network, called SegCaps. We extend the masked reconstruction regularization to the task of segmentation and perform thorough ablation experiments on each component of our method. The proposed convolutional-deconvolutional capsule network, SegCaps, shows state-of-the-art results while using a fraction of the parameters of popular segmentation networks. To validate our proposed method, we perform experiments segmenting pathological lungs from clinical and pre-clinical thoracic computed tomography (CT) scans and segmenting muscle and adipose (fat) tissue from magnetic resonance imaging (MRI) scans of human subjects' thighs. Notably, our experiments in lung segmentation represent the largest-scale study in pathological lung segmentation in the literature, where we conduct experiments across five extremely challenging datasets, containing both clinical and pre-clinical subjects, and nearly 2000 computed-tomography scans. Our newly developed segmentation platform outperforms other methods across all datasets while utilizing less than 5% of the parameters in the popular U-Net for biomedical image segmentation. Further, we demonstrate capsules' ability to generalize to unseen rotations/reflections on natural images.
Transformer-based models are unable to process long sequences due to their self-attention operation, which scales quadratically with the sequence length. To address this limitation, we introduce the Longformer with an attention mechanism that scales linearly with sequence length, making it easy to process documents of thousands of tokens or longer. Longformer's attention mechanism is a drop-in replacement for the standard self-attention and combines a local windowed attention with a task motivated global attention. Following prior work on long-sequence transformers, we evaluate Longformer on character-level language modeling and achieve state-of-the-art results on text8 and enwik8. In contrast to most prior work, we also pretrain Longformer and finetune it on a variety of downstream tasks. Our pretrained Longformer consistently outperforms RoBERTa on long document tasks and sets new state-of-the-art results on WikiHop and TriviaQA. We finally introduce the Longformer-Encoder-Decoder (LED), a Longformer variant for supporting long document generative sequence-to-sequence tasks, and demonstrate its effectiveness on the arXiv summarization dataset.
In its most elementary form, compressed sensing studies the design of decoding algorithms to recover a sufficiently sparse vector or code from a lower dimensional linear measurement vector. Typically it is assumed that the decoder has access to the encoder matrix, which in the combinatorial case is sparse and binary. In this paper we consider the problem of designing a decoder to recover a set of sparse codes from their linear measurements alone, that is without access to encoder matrix. To this end we study the matrix factorisation task of recovering both the encoder and sparse coding matrices from the associated linear measurement matrix. The contribution of this paper is a computationally efficient decoding algorithm, Decoder-Expander Based Factorisation, with strong performance guarantees. In particular, under mild assumptions on the sparse coding matrix and by deploying a novel random encoder matrix, we prove that Decoder-Expander Based Factorisation recovers both the encoder and sparse coding matrix at the optimal measurement rate with high probability and from a near optimal number of measurement vectors. In addition, our experiments demonstrate the efficacy and computational efficiency of our algorithm in practice. Beyond compressed sensing our results may be of interest for researchers working in areas such as linear sketching, coding theory and matrix compression.
This paper addresses the problem of novel view synthesis by means of neural rendering, where we are interested in predicting the novel view at an arbitrary camera pose based on a given set of input images from other viewpoints. Using the known query pose and input poses, we create an ordered set of observations that leads to the target view. Thus, the problem of single novel view synthesis is reformulated as a sequential view prediction task. In this paper, the proposed Transformer-based Generative Query Network (T-GQN) extends the neural-rendering methods by adding two new concepts. First, we use multi-view attention learning between context images to obtain multiple implicit scene representations. Second, we introduce a sequential rendering decoder to predict an image sequence, including the target view, based on the learned representations. Finally, we evaluate our model on various challenging datasets and demonstrate that our model not only gives consistent predictions but also doesn't require any retraining for finetuning.
Large pre-trained language models (LMs) are known to encode substantial amounts of linguistic information. However, high-level reasoning skills, such as numerical reasoning, are difficult to learn from a language-modeling objective only. Consequently, existing models for numerical reasoning have used specialized architectures with limited flexibility. In this work, we show that numerical reasoning is amenable to automatic data generation, and thus one can inject this skill into pre-trained LMs, by generating large amounts of data, and training in a multi-task setup. We show that pre-training our model, GenBERT, on this data, dramatically improves performance on DROP (49.3 $\rightarrow$ 72.3 F1), reaching performance that matches state-of-the-art models of comparable size, while using a simple and general-purpose encoder-decoder architecture. Moreover, GenBERT generalizes well to math word problem datasets, while maintaining high performance on standard RC tasks. Our approach provides a general recipe for injecting skills into large pre-trained LMs, whenever the skill is amenable to automatic data augmentation.
Attention mechanism plays a dominant role in the sequence generation models and has been used to improve the performance of machine translation and abstractive text summarization. Different from neural machine translation, in the task of text summarization, salience estimation for words, phrases or sentences is a critical component, since the output summary is a distillation of the input text. Although the typical attention mechanism can conduct text fragment selection from the input text conditioned on the decoder states, there is still a gap to conduct direct and effective salience detection. To bring back direct salience estimation for summarization with neural networks, we propose a Multi-Attention Learning framework which contains two new attention learning components for salience estimation: supervised attention learning and unsupervised attention learning. We regard the attention weights as the salience information, which means that the semantic units with large attention value will be more important. The context information obtained based on the estimated salience is incorporated with the typical attention mechanism in the decoder to conduct summary generation. Extensive experiments on some benchmark datasets in different languages demonstrate the effectiveness of the proposed framework for the task of abstractive summarization.
News headline generation aims to produce a short sentence to attract readers to read the news. One news article often contains multiple keyphrases that are of interest to different users, which can naturally have multiple reasonable headlines. However, most existing methods focus on the single headline generation. In this paper, we propose generating multiple headlines with keyphrases of user interests, whose main idea is to generate multiple keyphrases of interest to users for the news first, and then generate multiple keyphrase-relevant headlines. We propose a multi-source Transformer decoder, which takes three sources as inputs: (a) keyphrase, (b) keyphrase-filtered article, and (c) original article to generate keyphrase-relevant, high-quality, and diverse headlines. Furthermore, we propose a simple and effective method to mine the keyphrases of interest in the news article and build a first large-scale keyphrase-aware news headline corpus, which contains over 180K aligned triples of $<$news article, headline, keyphrase$>$. Extensive experimental comparisons on the real-world dataset show that the proposed method achieves state-of-the-art results in terms of quality and diversity
This paper presents online-capable deep learning model for probabilistic vehicle trajectory prediction. We propose a simple encoder-decoder architecture based on multi-head attention. The proposed model generates the distribution of the predicted trajectories for multiple vehicles in parallel. Our approach to model the interactions can learn to attend to a few influential vehicles in an unsupervised manner, which can improve the interpretability of the network. The experiments using naturalistic trajectories at highway show the clear improvement in terms of positional error on both longitudinal and lateral direction.
Cross-modal hashing facilitates mapping of heterogeneous multimedia data into a common Hamming space, which can beutilized for fast and flexible retrieval across different modalities. In this paper, we propose a novel cross-modal hashingarchitecture-deep neural decoder cross-modal hashing (DNDCMH), which uses a binary vector specifying the presence of certainfacial attributes as an input query to retrieve relevant face images from a database. The DNDCMH network consists of two separatecomponents: an attribute-based deep cross-modal hashing (ADCMH) module, which uses a margin (m)-based loss function toefficiently learn compact binary codes to preserve similarity between modalities in the Hamming space, and a neural error correctingdecoder (NECD), which is an error correcting decoder implemented with a neural network. The goal of NECD network in DNDCMH isto error correct the hash codes generated by ADCMH to improve the retrieval efficiency. The NECD network is trained such that it hasan error correcting capability greater than or equal to the margin (m) of the margin-based loss function. This results in NECD cancorrect the corrupted hash codes generated by ADCMH up to the Hamming distance of m. We have evaluated and comparedDNDCMH with state-of-the-art cross-modal hashing methods on standard datasets to demonstrate the superiority of our method.
This paper proposes a novel U-Net variant using stacked dilated convolutions for medical image segmentation (SDU-Net). SDU-Net adopts the architecture of vanilla U-Net with modifications in the encoder and decoder operations (an operation indicates all the processing for feature maps of the same resolution). Unlike vanilla U-Net which incorporates two standard convolutions in each encoder/decoder operation, SDU-Net uses one standard convolution followed by multiple dilated convolutions and concatenates all dilated convolution outputs as input to the next operation. Experiments showed that SDU-Net outperformed vanilla U-Net, attention U-Net (AttU-Net), and recurrent residual U-Net (R2U-Net) in all four tested segmentation tasks while using parameters around 40% of vanilla U-Net's, 17% of AttU-Net's, and 15% of R2U-Net's.
Systematic financial trading strategies account for over 80% of trade volume in equities and a large chunk of the foreign exchange market. In spite of the availability of data from multiple markets, current approaches in trading rely mainly on learning trading strategies per individual market. In this paper, we take a step towards developing fully end-to-end global trading strategies that leverage systematic trends to produce superior market-specific trading strategies. We introduce QuantNet: an architecture that learns market-agnostic trends and use these to learn superior market-specific trading strategies. Each market-specific model is composed of an encoder-decoder pair. The encoder transforms market-specific data into an abstract latent representation that is processed by a global model shared by all markets, while the decoder learns a market-specific trading strategy based on both local and global information from the market-specific encoder and the global model. QuantNet uses recent advances in transfer and meta-learning, where market-specific parameters are free to specialize on the problem at hand, whilst market-agnostic parameters are driven to capture signals from all markets. By integrating over idiosyncratic market data we can learn general transferable dynamics, avoiding the problem of overfitting to produce strategies with superior returns. We evaluate QuantNet on historical data across 3103 assets in 58 global equity markets. Against the top performing baseline, QuantNet yielded 51% higher Sharpe and 69% Calmar ratios. In addition we show the benefits of our approach over the non-transfer learning variant, with improvements of 15% and 41% in Sharpe and Calmar ratios. Code available in appendix.
Neural summarization models suffer from the fixed-size input limitation: if text length surpasses the model's maximal number of input tokens, some document content (possibly summary-relevant) gets truncated Independently summarizing windows of maximal input size disallows for information flow between windows and leads to incoherent summaries. We propose windowing models for neural abstractive summarization of (arbitrarily) long texts. We extend the sequence-to-sequence model augmented with pointer generator network by (1) allowing the encoder to slide over different windows of the input document and (2) sharing the decoder and retaining its state across different input windows. We explore two windowing variants: Static Windowing precomputes the number of tokens the decoder should generate from each window (based on training corpus statistics); in Dynamic Windowing the decoder learns to emit a token that signals encoder's shift to the next input window. Empirical results render our models effective in their intended use-case: summarizing long texts with relevant content not bound to the very document beginning.
We have seen significant improvements in machine translation due to the usage of deep learning. While the improvements in translation quality are impressive, the encoder-decoder architecture enables many more possibilities. In this paper, we explore one of these, the generation of constraint translation. We focus on length constraints, which are essential if the translation should be displayed in a given format. In this work, we propose an end-to-end approach for this task. Compared to a traditional method that first translates and then performs sentence compression, the text compression is learned completely unsupervised. By combining the idea with zero-shot multilingual machine translation, we are also able to perform unsupervised monolingual sentence compression. In order to fulfill the length constraints, we investigated several methods to integrate the constraints into the model. Using the presented technique, we are able to significantly improve the translation quality under constraints. Furthermore, we are able to perform unsupervised monolingual sentence compression.
Telementoring surgeons as they perform surgery can be essential in the treatment of patients when in situ expertise is not available. Nonetheless, expert mentors are often unavailable to provide trainees with real-time medical guidance. When mentors are unavailable, a fallback autonomous mechanism should provide medical practitioners with the required guidance. However, AI/autonomous mentoring in medicine has been limited by the availability of generalizable prediction models, and surgical procedures datasets to train those models with. This work presents the initial steps towards the development of an intelligent artificial system for autonomous medical mentoring. Specifically, we present the first Database for AI Surgical Instruction (DAISI). DAISI leverages on images and instructions to provide step-by-step demonstrations of how to perform procedures from various medical disciplines. The dataset was acquired from real surgical procedures and data from academic textbooks. We used DAISI to train an encoder-decoder neural network capable of predicting medical instructions given a current view of the surgery. Afterwards, the instructions predicted by the network were evaluated using cumulative BLEU scores and input from expert physicians. According to the BLEU scores, the predicted and ground truth instructions were as high as 67% similar. Additionally, expert physicians subjectively assessed the algorithm using Likert scale, and considered that the predicted descriptions were related to the images. This work provides a baseline for AI algorithms to assist in autonomous medical mentoring.
We introduce a simple and effective network architecture for monocular 3D hand pose estimation consisting of an image encoder followed by a mesh convolutional decoder that is trained through a direct 3D hand mesh reconstruction loss. We train our network by gathering a large-scale dataset of hand action in YouTube videos and use it as a source of weak supervision. Our weakly-supervised mesh convolutions-based system largely outperforms state-of-the-art methods, even halving the errors on the in the wild benchmark. The dataset and additional resources are available at https://arielai.com/mesh_hands.
We propose a novel method for hierarchical entity classification that embraces ontological structure at both training and during prediction. At training, our novel multi-level learning-to-rank loss compares positive types against negative siblings according to the type tree. During prediction, we define a coarse-to-fine decoder that restricts viable candidates at each level of the ontology based on already predicted parent type(s). We achieve state-of-the-art across multiple datasets, particularly with respect to strict accuracy.
City models and height maps of urban areas serve as a valuable data source for numerous applications, such as disaster management or city planning. While this information is not globally available, it can be substituted by digital surface models (DSMs), automatically produced from inexpensive satellite imagery. However, stereo DSMs often suffer from noise and blur. Furthermore, they are heavily distorted by vegetation, which is of lesser relevance for most applications. Such basic models can be filtered by convolutional neural networks (CNNs), trained on labels derived from digital elevation models (DEMs) and 3D city models, in order to obtain a refined DSM. We propose a modular multi-task learning concept that consolidates existing approaches into a generalized framework. Our encoder-decoder models with shared encoders and multiple task-specific decoders leverage roof type classification as a secondary task and multiple objectives including a conditional adversarial term. The contributing single-objective losses are automatically weighted in the final multi-task loss function based on learned uncertainty estimates. We evaluated the performance of specific instances of this family of network architectures. Our method consistently outperforms the state of the art on common data, both quantitatively and qualitatively, and generalizes well to a new dataset of an independent study area.
Although Shannon theory states that it is asymptotically optimal to separate the source and channel coding as two independent processes, in many practical communication scenarios this decomposition is limited by the finite bit-length and computational power for decoding. Recently, neural joint source-channel coding (NECST) is proposed to sidestep this problem. While it leverages the advancements of amortized inference and deep learning to improve the encoding and decoding process, it still cannot always achieve compelling results in terms of compression and error correction performance due to the limited robustness of its learned coding networks. In this paper, motivated by the inherent connections between neural joint source-channel coding and discrete representation learning, we propose a novel regularization method called Infomax Adversarial-Bit-Flip (IABF) to improve the stability and robustness of the neural joint source-channel coding scheme. More specifically, on the encoder side, we propose to explicitly maximize the mutual information between the codeword and data; while on the decoder side, the amortized reconstruction is regularized within an adversarial framework. Extensive experiments conducted on various real-world datasets evidence that our IABF can achieve state-of-the-art performances on both compression and error correction benchmarks and outperform the baselines by a significant margin.