PyTorchKR Papers

최신 논문

총 9,642편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1912.05122공식 코드 · Papers with Code

Towards Better Forecasting by Fusing Near and Distant Future Visions

Multivariate time series forecasting is an important yet challenging problem in machine learning. Most existing approaches only forecast the series value of one future moment, ignoring the interactions between predictions of future moments with different temporal distance. Such a deficiency probably prevents the model from getting enough information about the future, thus limiting the forecasting accuracy. To address this problem, we propose Multi-Level Construal Neural Network (MLCNN), a novel multi-task deep learning framework. Inspired by the Construal Level Theory of psychology, this model aims to improve the predictive performance by fusing forecasting information (i.e., future visions) of different future time. We first use the Convolution Neural Network to extract multi-level abstract representations of the raw data for near and distant future predictions. We then model the interplay between multiple predictive tasks and fuse their future visions through a modified Encoder-Decoder architecture. Finally, we combine traditional Autoregression model with the neural network to solve the scale insensitive problem. Experiments on three real-world datasets show that our method achieves statistically significant improvements compared to the most state-of-the-art baseline methods, with average 4.59% reduction on RMSE metric and average 6.87% reduction on MAE metric.

arXiv:1912.05074공식 코드 · Papers with Code

UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation

The state-of-the-art models for medical image segmentation are variants of U-Net and fully convolutional networks (FCN). Despite their success, these models have two limitations: (1) their optimal depth is apriori unknown, requiring extensive architecture search or inefficient ensemble of models of varying depths; and (2) their skip connections impose an unnecessarily restrictive fusion scheme, forcing aggregation only at the same-scale feature maps of the encoder and decoder sub-networks. To overcome these two limitations, we propose UNet++, a new neural architecture for semantic and instance segmentation, by (1) alleviating the unknown network depth with an efficient ensemble of U-Nets of varying depths, which partially share an encoder and co-learn simultaneously using deep supervision; (2) redesigning skip connections to aggregate features of varying semantic scales at the decoder sub-networks, leading to a highly flexible feature fusion scheme; and (3) devising a pruning scheme to accelerate the inference speed of UNet++. We have evaluated UNet++ using six different medical image segmentation datasets, covering multiple imaging modalities such as computed tomography (CT), magnetic resonance imaging (MRI), and electron microscopy (EM), and demonstrating that (1) UNet++ consistently outperforms the baseline models for the task of semantic segmentation across different datasets and backbone architectures; (2) UNet++ enhances segmentation quality of varying-size objects -- an improvement over the fixed-depth U-Net; (3) Mask RCNN++ (Mask R-CNN with UNet++ design) outperforms the original Mask R-CNN for the task of instance segmentation; and (4) pruned UNet++ models achieve significant speedup while showing only modest performance degradation. Our implementation and pre-trained models are available at https://github.com/MrGiovanni/UNetPlusPlus.

arXiv:1912.03761공식 코드 · Papers with Code

Variationally Regularized Graph-based Representation Learning for Electronic Health Records

Electronic Health Records (EHR) are high-dimensional data with implicit connections among thousands of medical concepts. These connections, for instance, the co-occurrence of diseases and lab-disease correlations can be informative when only a subset of these variables is documented by the clinician. A feasible approach to improving the representation learning of EHR data is to associate relevant medical concepts and utilize these connections. Existing medical ontologies can be the reference for EHR structures, but they place numerous constraints on the data source. Recent progress on graph neural networks (GNN) enables end-to-end learning of topological structures for non-grid or non-sequential data. However, there are problems to be addressed on how to learn the medical graph adaptively and how to understand the effect of the medical graph on representation learning. In this paper, we propose a variationally regularized encoder-decoder graph network that achieves more robustness in graph structure learning by regularizing node representations. Our model outperforms the existing graph and non-graph based methods in various EHR predictive tasks based on both public data and real-world clinical data. Besides the improvements in empirical experiment performances, we provide an interpretation of the effect of variational regularization compared to standard graph neural network, using singular value analysis.

arXiv:1912.03656공식 코드 · Papers with Code

Bidirectional Scene Text Recognition with a Single Decoder

Scene Text Recognition (STR) is the problem of recognizing the correct word or character sequence in a cropped word image. To obtain more robust output sequences, the notion of bidirectional STR has been introduced. So far, bidirectional STRs have been implemented by using two separate decoders; one for left-to-right decoding and one for right-to-left. Having two separate decoders for almost the same task with the same output space is undesirable from a computational and optimization point of view. We introduce the bidirectional Scene Text Transformer (Bi-STET), a novel bidirectional STR method with a single decoder for bidirectional text decoding. With its single decoder, Bi-STET outperforms methods that apply bidirectional decoding by using two separate decoders while also being more efficient than those methods, Furthermore, we achieve or beat state-of-the-art (SOTA) methods on all STR benchmarks with Bi-STET. Finally, we provide analyses and insights into the performance of Bi-STET.

arXiv:1912.03641

SaLite : A light-weight model for salient object detection

Salient object detection is a prevalent computer vision task that has applications ranging from abnormality detection to abnormality processing. Context modelling is an important criterion in the domain of saliency detection. A global context helps in determining the salient object in a given image by contrasting away other objects in the global view of the scene. However, the local context features detects the boundaries of the salient object with higher accuracy in a given region. To incorporate the best of both worlds, our proposed SaLite model uses both global and local contextual features. It is an encoder-decoder based architecture in which the encoder uses a lightweight SqueezeNet and decoder is modelled using convolution layers. Modern deep based models entitled for saliency detection use a large number of parameters, which is difficult to deploy on embedded systems. This paper attempts to solve the above problem using SaLite which is a lighter process for salient object detection without compromising on performance. Our approach is extensively evaluated on three publicly available datasets namely DUTS, MSRA10K, and SOC. Experimental results show that our proposed SaLite has significant and consistent improvements over the state-of-the-art methods.

arXiv:1912.04063

Goal-Conditioned Variational Autoencoder Trajectory Primitives with Continuous and Discrete Latent Codes

Imitation learning is an intuitive approach for teaching motion to robotic systems. Although previous studies have proposed various methods to model demonstrated movement primitives, one of the limitations of existing methods is that the shape of the trajectories are encoded in high dimensional space. The high dimensionality of the trajectory representation can be a bottleneck in the subsequent process such as planning a sequence of primitive motions. We address this problem by learning the latent space of the robot trajectory. If the latent variable of the trajectories can be learned, it can be used to tune the trajectory in an intuitive manner even when the user is not an expert. We propose a framework for modeling demonstrated trajectories with a neural network that learns the low-dimensional latent space. Our neural network structure is built on the variational autoencoder (VAE) with discrete and continuous latent variables. We extend the structure of the existing VAE to obtain the decoder that is conditioned on the goal position of the trajectory for generalization to different goal positions. Although the inference performed by VAE is not accurate, the positioning error at the generalized goal position can be reduced to less than 1~mm by incorporating the projection onto the solution space. To cope with requirement of the massive training data, we use a trajectory augmentation technique inspired by the data augmentation commonly used in the computer vision community. In the proposed framework, the latent variables that encodes the multiple types of trajectories are learned in an unsupervised manner, although existing methods usually require label information to model diverse behaviors. The learned decoder can be used as a motion planner in which the user can specify the goal position and the trajectory types by setting the latent variables.

arXiv:1912.02958

Synchronous Transformers for End-to-End Speech Recognition

For most of the attention-based sequence-to-sequence models, the decoder predicts the output sequence conditioned on the entire input sequence processed by the encoder. The asynchronous problem between the encoding and decoding makes these models difficult to be applied for online speech recognition. In this paper, we propose a model named synchronous transformer to address this problem, which can predict the output sequence chunk by chunk. Once a fixed-length chunk of the input sequence is processed by the encoder, the decoder begins to predict symbols immediately. During training, a forward-backward algorithm is introduced to optimize all the possible alignment paths. Our model is evaluated on a Mandarin dataset AISHELL-1. The experiments show that the synchronous transformer is able to perform encoding and decoding synchronously, and achieves a character error rate of 8.91% on the test set.

arXiv:1912.02914

RED-NET: A Recursive Encoder-Decoder Network for Edge Detection

In this paper, we introduce RED-NET: A Recursive Encoder-Decoder Network with Skip-Connections for edge detection in natural images. The proposed network is a novel integration of a Recursive Neural Network with an Encoder-Decoder architecture. The recursive network enables us to increase the network depth without increasing the number of parameters. Adding skip-connections between encoder and decoder helps the gradients reach all the layers of a network more easily and allows information related to finer details in the early stage of the encoder to be fully utilized in the decoder. Based on our extensive experiments on popular boundary detection datasets including BSDS500 \cite{Arbelaez2011}, NYUD \cite{Silberman2012} and Pascal Context \cite{Mottaghi2014}, RED-NET significantly advances the state-of-the-art on edge detection regarding standard evaluation metrics such as Optimal Dataset Scale (ODS) F-measure, Optimal Image Scale (OIS) F-measure, and Average Precision (AP).

arXiv:1912.02893

Learning undirected models via query training

Typical amortized inference in variational autoencoders is specialized for a single probabilistic query. Here we propose an inference network architecture that generalizes to unseen probabilistic queries. Instead of an encoder-decoder pair, we can train a single inference network directly from data, using a cost function that is stochastic not only over samples, but also over queries. We can use this network to perform the same inference tasks as we would in an undirected graphical model with hidden variables, without having to deal with the intractable partition function. The results can be mapped to the learning of an actual undirected model, which is a notoriously hard problem. Our network also marginalizes nuisance variables as required. We show that our approach generalizes to unseen probabilistic queries on also unseen test data, providing fast and flexible inference. Experiments show that this approach outperforms or matches PCD and AdVIL on 9 benchmark datasets.

arXiv:1912.03015

Learning to Correspond Dynamical Systems

Many dynamical systems exhibit similar structure, as often captured by hand-designed simplified models that can be used for analysis and control. We develop a method for learning to correspond pairs of dynamical systems via a learned latent dynamical system. Given trajectory data from two dynamical systems, we learn a shared latent state space and a shared latent dynamics model, along with an encoder-decoder pair for each of the original systems. With the learned correspondences in place, we can use a simulation of one system to produce an imagined motion of its counterpart. We can also simulate in the learned latent dynamics and synthesize the motions of both corresponding systems, as a form of bisimulation. We demonstrate the approach using pairs of controlled bipedal walkers, as well as by pairing a walker with a controlled pendulum.

arXiv:1912.03010공식 코드 · Papers with Code

Semantic Mask for Transformer based End-to-End Speech Recognition

Attention-based encoder-decoder model has achieved impressive results for both automatic speech recognition (ASR) and text-to-speech (TTS) tasks. This approach takes advantage of the memorization capacity of neural networks to learn the mapping from the input sequence to the output sequence from scratch, without the assumption of prior knowledge such as the alignments. However, this model is prone to overfitting, especially when the amount of training data is limited. Inspired by SpecAugment and BERT, in this paper, we propose a semantic mask based regularization for training such kind of end-to-end (E2E) model. The idea is to mask the input features corresponding to a particular output token, e.g., a word or a word-piece, in order to encourage the model to fill the token based on the contextual information. While this approach is applicable to the encoder-decoder framework with any type of neural network architecture, we study the transformer-based model for ASR in this work. We perform experiments on Librispeech 960h and TedLium2 data sets, and achieve the state-of-the-art performance on the test set in the scope of E2E models.

arXiv:1912.02738공식 코드 · Papers with Code

MetaFun: Meta-Learning with Iterative Functional Updates

We develop a functional encoder-decoder approach to supervised meta-learning, where labeled data is encoded into an infinite-dimensional functional representation rather than a finite-dimensional one. Furthermore, rather than directly producing the representation, we learn a neural update rule resembling functional gradient descent which iteratively improves the representation. The final representation is used to condition the decoder to make predictions on unlabeled data. Our approach is the first to demonstrates the success of encoder-decoder style meta-learning methods like conditional neural processes on large-scale few-shot classification benchmarks such as miniImageNet and tieredImageNet, where it achieves state-of-the-art performance.

arXiv:1912.02613

Singing Voice Conversion with Disentangled Representations of Singer and Vocal Technique Using Variational Autoencoders

We propose a flexible framework that deals with both singer conversion and singers vocal technique conversion. The proposed model is trained on non-parallel corpora, accommodates many-to-many conversion, and leverages recent advances of variational autoencoders. It employs separate encoders to learn disentangled latent representations of singer identity and vocal technique separately, with a joint decoder for reconstruction. Conversion is carried out by simple vector arithmetic in the learned latent spaces. Both a quantitative analysis as well as a visualization of the converted spectrograms show that our model is able to disentangle singer identity and vocal technique and successfully perform conversion of these attributes. To the best of our knowledge, this is the first work to jointly tackle conversion of singer identity and vocal technique based on a deep learning approach.

arXiv:1912.02288공식 코드 · Papers with Code

Simplified Action Decoder for Deep Multi-Agent Reinforcement Learning

In recent years we have seen fast progress on a number of benchmark problems in AI, with modern methods achieving near or super human performance in Go, Poker and Dota. One common aspect of all of these challenges is that they are by design adversarial or, technically speaking, zero-sum. In contrast to these settings, success in the real world commonly requires humans to collaborate and communicate with others, in settings that are, at least partially, cooperative. In the last year, the card game Hanabi has been established as a new benchmark environment for AI to fill this gap. In particular, Hanabi is interesting to humans since it is entirely focused on theory of mind, i.e., the ability to effectively reason over the intentions, beliefs and point of view of other agents when observing their actions. Learning to be informative when observed by others is an interesting challenge for Reinforcement Learning (RL): Fundamentally, RL requires agents to explore in order to discover good policies. However, when done naively, this randomness will inherently make their actions less informative to others during training. We present a new deep multi-agent RL method, the Simplified Action Decoder (SAD), which resolves this contradiction exploiting the centralized training phase. During training SAD allows other agents to not only observe the (exploratory) action chosen, but agents instead also observe the greedy action of their team mates. By combining this simple intuition with best practices for multi-agent learning, SAD establishes a new SOTA for learning methods for 2-5 players on the self-play part of the Hanabi challenge. Our ablations show the contributions of SAD compared with the best practice components. All of our code and trained agents are available at https://github.com/facebookresearch/Hanabi_SAD.

arXiv:1912.01881

Better Understanding Hierarchical Visual Relationship for Image Caption

The Convolutional Neural Network (CNN) has been the dominant image feature extractor in computer vision for years. However, it fails to get the relationship between images/objects and their hierarchical interactions which can be helpful for representing and describing an image. In this paper, we propose a new design for image caption under a general encoder-decoder framework. It takes into account the hierarchical interactions between different abstraction levels of visual information in the images and their bounding-boxes. Specifically, we present CNN plus Graph Convolutional Network (GCN) architecture that novelly integrates both semantic and spatial visual relationships into image encoder. The representations of regions in an image and the connections between images are refined by leveraging graph structure through GCN. With the learned multi-level features, our model capitalizes on the Transformer-based decoder for description generation. We conduct experiments on the COCO image captioning dataset. Evaluations show that our proposed model outperforms the previous state-of-the-art models in the task of image caption, leading to a better performance in terms of all evaluation metrics.

arXiv:1912.01852

PitchNet: Unsupervised Singing Voice Conversion with Pitch Adversarial Network

Singing voice conversion is to convert a singer's voice to another one's voice without changing singing content. Recent work shows that unsupervised singing voice conversion can be achieved with an autoencoder-based approach [1]. However, the converted singing voice can be easily out of key, showing that the existing approach cannot model the pitch information precisely. In this paper, we propose to advance the existing unsupervised singing voice conversion method proposed in [1] to achieve more accurate pitch translation and flexible pitch manipulation. Specifically, the proposed PitchNet added an adversarially trained pitch regression network to enforce the encoder network to learn pitch invariant phoneme representation, and a separate module to feed pitch extracted from the source audio to the decoder network. Our evaluation shows that the proposed method can greatly improve the quality of the converted singing voice (2.92 vs 3.75 in MOS). We also demonstrate that the pitch of converted singing can be easily controlled during generation by changing the levels of the extracted pitch before passing it to the decoder network.

arXiv:1912.01682

AMR-to-Text Generation with Cache Transition Systems

Text generation from AMR involves emitting sentences that reflect the meaning of their AMR annotations. Neural sequence-to-sequence models have successfully been used to decode strings from flattened graphs (e.g., using depth-first or random traversal). Such models often rely on attention-based decoders to map AMR node to English token sequences. Instead of linearizing AMR, we directly encode its graph structure and delegate traversal to the decoder. To enforce a sentence-aligned graph traversal and provide local graph context, we predict transition-based parser actions in addition to English words. We present two model variants: one generates parser actions prior to words, while the other interleaves actions with words.

arXiv:1912.02047

Neural Machine Translation: A Review and Survey

The field of machine translation (MT), the automatic translation of written text from one natural language into another, has experienced a major paradigm shift in recent years. Statistical MT, which mainly relies on various count-based models and which used to dominate MT research for decades, has largely been superseded by neural machine translation (NMT), which tackles translation with a single neural network. In this work we will trace back the origins of modern NMT architectures to word and sentence embeddings and earlier examples of the encoder-decoder network family. We will conclude with a survey of recent trends in the field.

arXiv:1912.01456

Facial Expression Representation Learning by Synthesizing Expression Images

Representations used for Facial Expression Recognition (FER) usually contain expression information along with identity features. In this paper, we propose a novel Disentangled Expression learning-Generative Adversarial Network (DE-GAN) which combines the concept of disentangled representation learning with residue learning to explicitly disentangle facial expression representation from identity information. In this method the facial expression representation is learned by reconstructing an expression image employing an encoder-decoder based generator. Unlike previous works using only expression residual learning for facial expression recognition, our method learns the disentangled expression representation along with the expressive component recorded by the encoder of DE-GAN. In order to improve the quality of synthesized expression images and the effectiveness of the learned disentangled expression representation, expression and identity classification is performed by the discriminator of DE-GAN. Experiments performed on widely used datasets (CK+, MMI, Oulu-CASIA) show that the proposed technique produces comparable or better results than state-of-the-art methods.

arXiv:1912.01394

RGPNet: A Real-Time General Purpose Semantic Segmentation

We propose a real-time general purpose semantic segmentation architecture, RGPNet, which achieves significant performance gain in complex environments. RGPNet consists of a light-weight asymmetric encoder-decoder and an adaptor. The adaptor helps preserve and refine the abstract concepts from multiple levels of distributed representations between the encoder and decoder. It also facilitates the gradient flow from deeper layers to shallower layers. Our experiments demonstrate that RGPNet can generate segmentation results in real-time with comparable accuracy to the state-of-the-art non-real-time heavy models. Moreover, towards green AI, we show that using an optimized label-relaxation technique with progressive resizing can reduce the training time by up to 60% while preserving the performance. We conclude that RGPNet obtains a better speed-accuracy trade-off across multiple datasets.