Predicting unseen weather phenomena is an important issue for disaster management. In this paper, we suggest a model for a convolutional sequence-to-sequence autoencoder for predicting undiscovered weather situations from previous satellite images. We also propose a symmetric skip connection between encoder and decoder modules to produce more comprehensive image predictions. To examine our model performance, we conducted experiments for each suggested model to predict future satellite images from historical satellite images. A specific combination of skip connection and sequence-to-sequence autoencoder was able to generate closest prediction from the ground truth image.
We present an approach for identifying the most walkable direction for navigation using a hand-held camera. Our approach extracts semantically rich contextual information from the scene using a custom encoder-decoder architecture for semantic segmentation and models the spatial and temporal behavior of objects in the scene using a spatio-temporal graph. The system learns to minimize a cost function over the spatial and temporal object attributes to identify the most walkable direction. We construct a new annotated navigation dataset collected using a hand-held mobile camera in an unconstrained outdoor environment, which includes challenging settings such as highly dynamic scenes, occlusion between objects, and distortions. Our system achieves an accuracy of 84% on predicting a safe direction. We also show that our custom segmentation network is both fast and accurate, achieving mIOU (mean intersection over union) scores of 81 and 44.7 on the PASCAL VOC and the PASCAL Context datasets, respectively, while running at about 21 frames per second.
We apply sequence-to-sequence model to mitigate the impact of speech recognition errors on open domain end-to-end dialog generation. We cast the task as a domain adaptation problem where ASR transcriptions and original text are in two different domains. In this paper, our proposed model includes two individual encoders for each domain data and make their hidden states similar to ensure the decoder predict the same dialog text. The method shows that the sequence-to-sequence model can learn the ASR transcriptions and original text pair having the same meaning and eliminate the speech recognition errors. Experimental results on Cornell movie dialog dataset demonstrate that the domain adaption system help the spoken dialog system generate more similar responses with the original text answers.
In this study, we present a novel end-to-end approach based on the encoder-decoder framework with the attention mechanism for online handwritten mathematical expression recognition (OHMER). First, the input two-dimensional ink trajectory information of handwritten expression is encoded via the gated recurrent unit based recurrent neural network (GRU-RNN). Then the decoder is also implemented by the GRU-RNN with a coverage-based attention model. The proposed approach can simultaneously accomplish the symbol recognition and structural analysis to output a character sequence in LaTeX format. Validated on the CROHME 2014 competition task, our approach significantly outperforms the state-of-the-art with an expression recognition accuracy of 52.43% by only using the official training dataset. Furthermore, the alignments between the input trajectories of handwritten expressions and the output LaTeX sequences are visualized by the attention mechanism to show the effectiveness of the proposed method.
Information extraction from textual documents such as hospital records and healthrelated user discussions has become a topic of intense interest. The task of medical concept coding is to map a variable length text to medical concepts and corresponding classification codes in some external system or ontology. In this work, we utilize recurrent neural networks to automatically assign ICD-10 codes to fragments of death certificates written in English. We develop end-to-end neural architectures directly tailored to the task, including basic encoder-decoder architecture for statistical translation. In order to incorporate prior knowledge, we concatenate cosine similarities vector among the text and dictionary entry to the encoded state. Being applied to a standard benchmark from CLEF eHealth 2017 challenge, our model achieved F-measure of 85.01% on a full test set with significant improvement as compared to the average score of 62.2% for all official participants approaches.
Text attribute transfer using non-parallel data requires methods that can perform disentanglement of content and linguistic attributes. In this work, we propose multiple improvements over the existing approaches that enable the encoder-decoder framework to cope with the text attribute transfer from non-parallel data. We perform experiments on the sentiment transfer task using two datasets. For both datasets, our proposed method outperforms a strong baseline in two of the three employed evaluation metrics.
Unsupervised Image-to-Image Translation achieves spectacularly advanced developments nowadays. However, recent approaches mainly focus on one model with two domains, which may face heavy burdens with large cost of $O(n^2)$ training time and model parameters, under such a requirement that $n$ domains are freely transferred to each other in a general setting. To address this problem, we propose a novel and unified framework named Domain-Bank, which consists of a global shared auto-encoder and $n$ domain-specific encoders/decoders, assuming that a universal shared-latent sapce can be projected. Thus, we yield $O(n)$ complexity in model parameters along with a huge reduction of the time budgets. Besides the high efficiency, we show the comparable (or even better) image translation results over state-of-the-arts on various challenging unsupervised image translation tasks, including face image translation, fashion-clothes translation and painting style translation. We also apply the proposed framework to domain adaptation and achieve state-of-the-art performance on digit benchmark datasets. Further, thanks to the explicit representation of the domain-specific decoders as well as the universal shared-latent space, it also enables us to conduct incremental learning to add a new domain encoder/decoder. Linear combination of different domains' representations is also obtained by fusing the corresponding decoders.
We propose an approach for forecasting video of complex human activity involving multiple people. Direct pixel-level prediction is too simple to handle the appearance variability in complex activities. Hence, we develop novel intermediate representations. An architecture combining a hierarchical temporal model for predicting human poses and encoder-decoder convolutional neural networks for rendering target appearances is proposed. Our hierarchical model captures interactions among people by adopting a dynamic group-based interaction mechanism. Next, our appearance rendering network encodes the targets' appearances by learning adaptive appearance filters using a fully convolutional network. Finally, these filters are placed in encoder-decoder neural networks to complete the rendering. We demonstrate that our model can generate videos that are superior to state-of-the-art methods, and can handle complex human activity scenarios in video forecasting.
Facial attribute editing aims to manipulate single or multiple attributes of a face image, i.e., to generate a new face with desired attributes while preserving other details. Recently, generative adversarial net (GAN) and encoder-decoder architecture are usually incorporated to handle this task with promising results. Based on the encoder-decoder architecture, facial attribute editing is achieved by decoding the latent representation of the given face conditioned on the desired attributes. Some existing methods attempt to establish an attribute-independent latent representation for further attribute editing. However, such attribute-independent constraint on the latent representation is excessive because it restricts the capacity of the latent representation and may result in information loss, leading to over-smooth and distorted generation. Instead of imposing constraints on the latent representation, in this work we apply an attribute classification constraint to the generated image to just guarantee the correct change of desired attributes, i.e., to "change what you want". Meanwhile, the reconstruction learning is introduced to preserve attribute-excluding details, in other words, to "only change what you want". Besides, the adversarial learning is employed for visually realistic editing. These three components cooperate with each other forming an effective framework for high quality facial attribute editing, referred as AttGAN. Furthermore, our method is also directly applicable for attribute intensity control and can be naturally extended for attribute style manipulation. Experiments on CelebA dataset show that our method outperforms the state-of-the-arts on realistic attribute editing with facial details well preserved.
The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.
We report a method to convert discrete representations of molecules to and from a multidimensional continuous representation. This model allows us to generate new molecules for efficient exploration and optimization through open-ended spaces of chemical compounds. A deep neural network was trained on hundreds of thousands of existing chemical structures to construct three coupled functions: an encoder, a decoder and a predictor. The encoder converts the discrete representation of a molecule into a real-valued continuous vector, and the decoder converts these continuous vectors back to discrete molecular representations. The predictor estimates chemical properties from the latent continuous vector representation of the molecule. Continuous representations allow us to automatically generate novel chemical structures by performing simple operations in the latent space, such as decoding random vectors, perturbing known chemical structures, or interpolating between molecules. Continuous representations also allow the use of powerful gradient-based optimization to efficiently guide the search for optimized functional compounds. We demonstrate our method in the domain of drug-like molecules and also in the set of molecules with fewer that nine heavy atoms.
Learning a goal-oriented dialog policy is generally performed offline with supervised learning algorithms or online with reinforcement learning (RL). Additionally, as companies accumulate massive quantities of dialog transcripts between customers and trained human agents, encoder-decoder methods have gained popularity as agent utterances can be directly treated as supervision without the need for utterance-level annotations. However, one potential drawback of such approaches is that they myopically generate the next agent utterance without regard for dialog-level considerations. To resolve this concern, this paper describes an offline RL method for learning from unannotated corpora that can optimize a goal-oriented policy at both the utterance and dialog level. We introduce a novel reward function and use both on-policy and off-policy policy gradient to learn a policy offline without requiring online user interaction or an explicit state space definition.
Attention-based Encoder-Decoder has the effective architecture for neural machine translation (NMT), which typically relies on recurrent neural networks (RNN) to build the blocks that will be lately called by attentive reader during the decoding process. This design of encoder yields relatively uniform composition on source sentence, despite the gating mechanism employed in encoding RNN. On the other hand, we often hope the decoder to take pieces of source sentence at varying levels suiting its own linguistic structure: for example, we may want to take the entity name in its raw form while taking an idiom as a perfectly composed unit. Motivated by this demand, we propose Multi-channel Encoder (MCE), which enhances encoding components with different levels of composition. More specifically, in addition to the hidden state of encoding RNN, MCE takes 1) the original word embedding for raw encoding with no composition, and 2) a particular design of external memory in Neural Turing Machine (NTM) for more complex composition, while all three encoding strategies are properly blended during decoding. Empirical study on Chinese-English translation shows that our model can improve by 6.52 BLEU points upon a strong open source NMT system: DL4MT1. On the WMT14 English- French task, our single shallow system achieves BLEU=38.8, comparable with the state-of-the-art deep models.
In this work we propose a novel model-based deep convolutional autoencoder that addresses the highly challenging problem of reconstructing a 3D human face from a single in-the-wild color image. To this end, we combine a convolutional encoder network with an expert-designed generative model that serves as decoder. The core innovation is our new differentiable parametric decoder that encapsulates image formation analytically based on a generative model. Our decoder takes as input a code vector with exactly defined semantic meaning that encodes detailed face pose, shape, expression, skin reflectance and scene illumination. Due to this new way of combining CNN-based with model-based face reconstruction, the CNN-based encoder learns to extract semantically meaningful parameters from a single monocular input image. For the first time, a CNN encoder and an expert-designed generative model can be trained end-to-end in an unsupervised manner, which renders training on very large (unlabeled) real world data feasible. The obtained reconstructions compare favorably to current state-of-the-art approaches in terms of quality and richness of representation.
We review some of the most recent approaches to colorize gray-scale images using deep learning methods. Inspired by these, we propose a model which combines a deep Convolutional Neural Network trained from scratch with high-level features extracted from the Inception-ResNet-v2 pre-trained model. Thanks to its fully convolutional architecture, our encoder-decoder model can process images of any size and aspect ratio. Other than presenting the training results, we assess the "public acceptance" of the generated images by means of a user study. Finally, we present a carousel of applications on different types of images, such as historical photographs.
Direct acoustics-to-word (A2W) models in the end-to-end paradigm have received increasing attention compared to conventional sub-word based automatic speech recognition models using phones, characters, or context-dependent hidden Markov model states. This is because A2W models recognize words from speech without any decoder, pronunciation lexicon, or externally-trained language model, making training and decoding with such models simple. Prior work has shown that A2W models require orders of magnitude more training data in order to perform comparably to conventional models. Our work also showed this accuracy gap when using the English Switchboard-Fisher data set. This paper describes a recipe to train an A2W model that closes this gap and is at-par with state-of-the-art sub-word based models. We achieve a word error rate of 8.8%/13.9% on the Hub5-2000 Switchboard/CallHome test sets without any decoder or language model. We find that model initialization, training data order, and regularization have the most impact on the A2W model performance. Next, we present a joint word-character A2W model that learns to first spell the word and then recognize it. This model provides a rich output to the user instead of simple word hypotheses, making it especially useful in the case of words unseen or rarely-seen during training.
Neural Machine Translation models have replaced the conventional phrase based statistical translation methods since the former takes a generic, scalable, data-driven approach rather than relying on manual, hand-crafted features. The neural machine translation system is based on one neural network that is composed of two parts, one that is responsible for input language sentence and other part that handles the desired output language sentence. This model based on encoder-decoder architecture also takes as input the distributed representations of the source language which enriches the learnt dependencies and gives a warm start to the network. In this work, we transform Roman-Urdu to Urdu transliteration into sequence to sequence learning problem. To this end, we make the following contributions. We create the first ever parallel corpora of Roman-Urdu to Urdu, create the first ever distributed representation of Roman-Urdu and present the first neural machine translation model that transliterates text from Roman-Urdu to Urdu language. Our model has achieved the state-of-the-art results using BLEU as the evaluation metric. Precisely, our model is able to correctly predict sentences up to length 10 while achieving BLEU score of 48.6 on the test set. We are hopeful that our model and our results shall serve as the baseline for further work in the domain of neural machine translation for Roman-Urdu to Urdu using distributed representation.
In information retrieval, learning to rank constructs a machine-based ranking model which given a query, sorts the search results by their degree of relevance or importance to the query. Neural networks have been successfully applied to this problem, and in this paper, we propose an attention-based deep neural network which better incorporates different embeddings of the queries and search results with an attention-based mechanism. This model also applies a decoder mechanism to learn the ranks of the search results in a listwise fashion. The embeddings are trained with convolutional neural networks or the word2vec model. We demonstrate the performance of this model with image retrieval and text querying data sets.
Positron emission tomography (PET) is widely used in various clinical applications, including cancer diagnosis, heart disease and neuro disorders. The use of radioactive tracer in PET imaging raises concerns due to the risk of radiation exposure. To minimize this potential risk in PET imaging, efforts have been made to reduce the amount of radio-tracer usage. However, lowing dose results in low Signal-to-Noise-Ratio (SNR) and loss of information, both of which will heavily affect clinical diagnosis. Besides, the ill-conditioning of low-dose PET image reconstruction makes it a difficult problem for iterative reconstruction algorithms. Previous methods proposed are typically complicated and slow, yet still cannot yield satisfactory results at significantly low dose. Here, we propose a deep learning method to resolve this issue with an encoder-decoder residual deep network with concatenate skip connections. Experiments shows the proposed method can reconstruct low-dose PET image to a standard-dose quality with only two-hundredth dose. Different cost functions for training model are explored. Multi-slice input strategy is introduced to provide the network with more structural information and make it more robust to noise. Evaluation on ultra-low-dose clinical data shows that the proposed method can achieve better result than the state-of-the-art methods and reconstruct images with comparable quality using only 0.5% of the original regular dose.
The paper approaches the task of handwritten text recognition (HTR) with attentional encoder-decoder networks trained on sequences of characters, rather than words. We experiment on lines of text from popular handwriting datasets and compare different activation functions for the attention mechanism used for aligning image pixels and target characters. We find that softmax attention focuses heavily on individual characters, while sigmoid attention focuses on multiple characters at each step of the decoding. When the sequence alignment is one-to-one, softmax attention is able to learn a more precise alignment at each step of the decoding, whereas the alignment generated by sigmoid attention is much less precise. When a linear function is used to obtain attention weights, the model predicts a character by looking at the entire sequence of characters and performs poorly because it lacks a precise alignment between the source and target. Future research may explore HTR in natural scene images, since the model is capable of transcribing handwritten text without the need for producing segmentations or bounding boxes of text in images.