PyTorchKR Papers

최신 논문

총 9,642편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1604.03390공식 코드 · Papers with Code

Video Description using Bidirectional Recurrent Neural Networks

Although traditionally used in the machine translation field, the encoder-decoder framework has been recently applied for the generation of video and image descriptions. The combination of Convolutional and Recurrent Neural Networks in these models has proven to outperform the previous state of the art, obtaining more accurate video descriptions. In this work we propose pushing further this model by introducing two contributions into the encoding stage. First, producing richer image representations by combining object and location information from Convolutional Neural Networks and second, introducing Bidirectional Recurrent Neural Networks for capturing both forward and backward temporal relationships in the input frames.

arXiv:1612.04949

Recurrent Image Captioner: Describing Images with Spatial-Invariant Transformation and Attention Filtering

Along with the prosperity of recurrent neural network in modelling sequential data and the power of attention mechanism in automatically identify salient information, image captioning, a.k.a., image description, has been remarkably advanced in recent years. Nonetheless, most existing paradigms may suffer from the deficiency of invariance to images with different scaling, rotation, etc.; and effective integration of standalone attention to form a holistic end-to-end system. In this paper, we propose a novel image captioning architecture, termed Recurrent Image Captioner (\textbf{RIC}), which allows visual encoder and language decoder to coherently cooperate in a recurrent manner. Specifically, we first equip CNN-based visual encoder with a differentiable layer to enable spatially invariant transformation of visual signals. Moreover, we deploy an attention filter module (differentiable) between encoder and decoder to dynamically determine salient visual parts. We also employ bidirectional LSTM to preprocess sentences for generating better textual representations. Besides, we propose to exploit variational inference to optimize the whole architecture. Extensive experimental results on three benchmark datasets (i.e., Flickr8k, Flickr30k and MS COCO) demonstrate the superiority of our proposed architecture as compared to most of the state-of-the-art methods.

arXiv:1612.07454

How to Train Your Deep Neural Network with Dictionary Learning

Currently there are two predominant ways to train deep neural networks. The first one uses restricted Boltzmann machine (RBM) and the second one autoencoders. RBMs are stacked in layers to form deep belief network (DBN); the final representation layer is attached to the target to complete the deep neural network. Autoencoders are nested one inside the other to form stacked autoencoders; once the stcaked autoencoder is learnt the decoder portion is detached and the target attached to the deepest layer of the encoder to form the deep neural network. This work proposes a new approach to train deep neural networks using dictionary learning as the basic building block; the idea is to use the features from the shallower layer as inputs for training the next deeper layer. One can use any type of dictionary learning (unsupervised, supervised, discriminative etc.) as basic units till the pre-final layer. In the final layer one needs to use the label consistent dictionary learning formulation for classification. We compare our proposed framework with existing state-of-the-art deep learning techniques on benchmark problems; we are always within the top 10 results. In actual problems of age and gender classification, we are better than the best known techniques.

arXiv:1612.08534

Robust LSTM-Autoencoders for Face De-Occlusion in the Wild

Face recognition techniques have been developed significantly in recent years. However, recognizing faces with partial occlusion is still challenging for existing face recognizers which is heavily desired in real-world applications concerning surveillance and security. Although much research effort has been devoted to developing face de-occlusion methods, most of them can only work well under constrained conditions, such as all the faces are from a pre-defined closed set. In this paper, we propose a robust LSTM-Autoencoders (RLA) model to effectively restore partially occluded faces even in the wild. The RLA model consists of two LSTM components, which aims at occlusion-robust face encoding and recurrent occlusion removal respectively. The first one, named multi-scale spatial LSTM encoder, reads facial patches of various scales sequentially to output a latent representation, and occlusion-robustness is achieved owing to the fact that the influence of occlusion is only upon some of the patches. Receiving the representation learned by the encoder, the LSTM decoder with a dual channel architecture reconstructs the overall face and detects occlusion simultaneously, and by feat of LSTM, the decoder breaks down the task of face de-occlusion into restoring the occluded part step by step. Moreover, to minimize identify information loss and guarantee face recognition accuracy over recovered faces, we introduce an identity-preserving adversarial training scheme to further improve RLA. Extensive experiments on both synthetic and real datasets of faces with occlusion clearly demonstrate the effectiveness of our proposed RLA in removing different types of facial occlusion at various locations. The proposed method also provides significantly larger performance gain than other de-occlusion methods in promoting recognition performance over partially-occluded faces.

arXiv:1612.08510

Learning Non-Lambertian Object Intrinsics across ShapeNet Categories

We consider the non-Lambertian object intrinsic problem of recovering diffuse albedo, shading, and specular highlights from a single image of an object. We build a large-scale object intrinsics database based on existing 3D models in the ShapeNet database. Rendered with realistic environment maps, millions of synthetic images of objects and their corresponding albedo, shading, and specular ground-truth images are used to train an encoder-decoder CNN. Once trained, the network can decompose an image into the product of albedo and shading components, along with an additive specular component. Our CNN delivers accurate and sharp results in this classical inverse problem of computer vision, sharp details attributed to skip layer connections at corresponding resolutions from the encoder to the decoder. Benchmarked on our ShapeNet and MIT intrinsics datasets, our model consistently outperforms the state-of-the-art by a large margin. We train and test our CNN on different object categories. Perhaps surprising especially from the CNN classification perspective, our intrinsics CNN generalizes very well across categories. Our analysis shows that feature learning at the encoder stage is more crucial for developing a universal representation across categories. We apply our synthetic data trained model to images and videos downloaded from the internet, and observe robust and realistic intrinsics results. Quality non-Lambertian intrinsics could open up many interesting applications such as image-based albedo and specular editing.

arXiv:1612.01756

Video Ladder Networks

We present the Video Ladder Network (VLN) for efficiently generating future video frames. VLN is a neural encoder-decoder model augmented at all layers by both recurrent and feedforward lateral connections. At each layer, these connections form a lateral recurrent residual block, where the feedforward connection represents a skip connection and the recurrent connection represents the residual. Thanks to the recurrent connections, the decoder can exploit temporal summaries generated from all layers of the encoder. This way, the top layer is relieved from the pressure of modeling lower-level spatial and temporal details. Furthermore, we extend the basic version of VLN to incorporate ResNet-style residual blocks in the encoder and decoder, which help improving the prediction results. VLN is trained in self-supervised regime on the Moving MNIST dataset, achieving competitive results while having very simple structure and providing fast inference.

arXiv:1612.02482

Improving the Performance of Neural Machine Translation Involving Morphologically Rich Languages

The advent of the attention mechanism in neural machine translation models has improved the performance of machine translation systems by enabling selective lookup into the source sentence. In this paper, the efficiencies of translation using bidirectional encoder attention decoder models were studied with respect to translation involving morphologically rich languages. The English - Tamil language pair was selected for this analysis. First, the use of Word2Vec embedding for both the English and Tamil words improved the translation results by 0.73 BLEU points over the baseline RNNSearch model with 4.84 BLEU score. The use of morphological segmentation before word vectorization to split the morphologically rich Tamil words into their respective morphemes before the translation, caused a reduction in the target vocabulary size by a factor of 8. Also, this model (RNNMorph) improved the performance of neural machine translation by 7.05 BLEU points over the RNNSearch model used over the same corpus. Since the BLEU evaluation of the RNNMorph model might be unreliable due to an increase in the number of matching tokens per sentence, the performances of the translations were also compared by means of human evaluation metrics of adequacy, fluency and relative ranking. Further, the use of morphological segmentation also improved the efficacy of the attention mechanism.

arXiv:1701.02133

Deep driven fMRI decoding of visual categories

Deep neural networks have been developed drawing inspiration from the brain visual pathway, implementing an end-to-end approach: from image data to video object classes. However building an fMRI decoder with the typical structure of Convolutional Neural Network (CNN), i.e. learning multiple level of representations, seems impractical due to lack of brain data. As a possible solution, this work presents the first hybrid fMRI and deep features decoding approach: collected fMRI and deep learnt representations of video object classes are linked together by means of Kernel Canonical Correlation Analysis. In decoding, this allows exploiting the discriminatory power of CNN by relating the fMRI representation to the last layer of CNN (fc7). We show the effectiveness of embedding fMRI data onto a subspace related to deep features in distinguishing semantic visual categories based solely on brain imaging data.

arXiv:1701.02073

Neural Personalized Response Generation as Domain Adaptation

In this paper, we focus on the personalized response generation for conversational systems. Based on the sequence to sequence learning, especially the encoder-decoder framework, we propose a two-phase approach, namely initialization then adaptation, to model the responding style of human and then generate personalized responses. For evaluation, we propose a novel human aided method to evaluate the performance of the personalized response generation models by online real-time conversation and offline human judgement. Moreover, the lexical divergence of the responses generated by the 5 personalized models indicates that the proposed two-phase approach achieves good results on modeling the responding style of human and generating personalized responses for the conversational systems.

arXiv:1701.04949

A Deep Convolutional Auto-Encoder with Pooling - Unpooling Layers in Caffe

This paper presents the development of several models of a deep convolutional auto-encoder in the Caffe deep learning framework and their experimental evaluation on the example of MNIST dataset. We have created five models of a convolutional auto-encoder which differ architecturally by the presence or absence of pooling and unpooling layers in the auto-encoder's encoder and decoder parts. Our results show that the developed models provide very good results in dimensionality reduction and unsupervised clustering tasks, and small classification errors when we used the learned internal code as an input of a supervised linear classifier and multi-layer perceptron. The best results were provided by a model where the encoder part contains convolutional and pooling layers, followed by an analogous decoder part with deconvolution and unpooling layers without the use of switch variables in the decoder part. The paper also discusses practical details of the creation of a deep convolutional auto-encoder in the very popular Caffe deep learning framework. We believe that our approach and results presented in this paper could help other researchers to build efficient deep neural network architectures in the future.

arXiv:1701.06521

Incorporating Global Visual Features into Attention-Based Neural Machine Translation

We introduce multi-modal, attention-based neural machine translation (NMT) models which incorporate visual features into different parts of both the encoder and the decoder. We utilise global image features extracted using a pre-trained convolutional neural network and incorporate them (i) as words in the source sentence, (ii) to initialise the encoder hidden state, and (iii) as additional data to initialise the decoder hidden state. In our experiments, we evaluate how these different strategies to incorporate global image features compare and which ones perform best. We also study the impact that adding synthetic multi-modal, multilingual data brings and find that the additional data have a positive impact on multi-modal models. We report new state-of-the-art results and our best models also significantly improve on a comparable phrase-based Statistical MT (PBSMT) model trained on the Multi30k data set according to all metrics evaluated. To the best of our knowledge, it is the first time a purely neural model significantly improves over a PBSMT model on all metrics evaluated on this data set.

arXiv:1612.06027

Neural Multi-Source Morphological Reinflection

We explore the task of multi-source morphological reinflection, which generalizes the standard, single-source version. The input consists of (i) a target tag and (ii) multiple pairs of source form and source tag for a lemma. The motivation is that it is beneficial to have access to more than one source form since different source forms can provide complementary information, e.g., different stems. We further present a novel extension to the encoder- decoder recurrent neural architecture, consisting of multiple encoders, to better solve the task. We show that our new architecture outperforms single-source reinflection models and publish our dataset for multi-source morphological reinflection to facilitate future research.

arXiv:1701.08198

Adversarial Evaluation of Dialogue Models

The recent application of RNN encoder-decoder models has resulted in substantial progress in fully data-driven dialogue systems, but evaluation remains a challenge. An adversarial loss could be a way to directly evaluate the extent to which generated dialogue responses sound like they came from a human. This could reduce the need for human evaluation, while more directly evaluating on a generative task. In this work, we investigate this idea by training an RNN to discriminate a dialogue model's samples from human-generated samples. Although we find some evidence this setup could be viable, we also note that many issues remain in its practical application. We discuss both aspects and conclude that future work is warranted.

arXiv:1701.07213

Learning from Label Proportions in Brain-Computer Interfaces: Online Unsupervised Learning with Guarantees

Objective: Using traditional approaches, a Brain-Computer Interface (BCI) requires the collection of calibration data for new subjects prior to online use. Calibration time can be reduced or eliminated e.g.~by transfer of a pre-trained classifier or unsupervised adaptive classification methods which learn from scratch and adapt over time. While such heuristics work well in practice, none of them can provide theoretical guarantees. Our objective is to modify an event-related potential (ERP) paradigm to work in unison with the machine learning decoder to achieve a reliable calibration-less decoding with a guarantee to recover the true class means. Method: We introduce learning from label proportions (LLP) to the BCI community as a new unsupervised, and easy-to-implement classification approach for ERP-based BCIs. The LLP estimates the mean target and non-target responses based on known proportions of these two classes in different groups of the data. We modified a visual ERP speller to meet the requirements of the LLP. For evaluation, we ran simulations on artificially created data sets and conducted an online BCI study with N=13 subjects performing a copy-spelling task. Results: Theoretical considerations show that LLP is guaranteed to minimize the loss function similarly to a corresponding supervised classifier. It performed well in simulations and in the online application, where 84.5% of characters were spelled correctly on average without prior calibration. Significance: The continuously adapting LLP classifier is the first unsupervised decoder for ERP BCIs guaranteed to find the true class means. This makes it an ideal solution to avoid a tedious calibration and to tackle non-stationarities in the data. Additionally, LLP works on complementary principles compared to existing unsupervised methods, allowing for their further enhancement when combined with LLP.

arXiv:1702.00307

Pixel-wise Ear Detection with Convolutional Encoder-Decoder Networks

Object detection and segmentation represents the basis for many tasks in computer and machine vision. In biometric recognition systems the detection of the region-of-interest (ROI) is one of the most crucial steps in the overall processing pipeline, significantly impacting the performance of the entire recognition system. Existing approaches to ear detection, for example, are commonly susceptible to the presence of severe occlusions, ear accessories or variable illumination conditions and often deteriorate in their performance if applied on ear images captured in unconstrained settings. To address these shortcomings, we present in this paper a novel ear detection technique based on convolutional encoder-decoder networks (CEDs). For our technique, we formulate the problem of ear detection as a two-class segmentation problem and train a convolutional encoder-decoder network based on the SegNet architecture to distinguish between image-pixels belonging to either the ear or the non-ear class. The output of the network is then post-processed to further refine the segmentation result and return the final locations of the ears in the input image. Different from competing techniques from the literature, our approach does not simply return a bounding box around the detected ear, but provides detailed, pixel-wise information about the location of the ears in the image. Our experiments on a dataset gathered from the web (a.k.a. in the wild) show that the proposed technique ensures good detection results in the presence of various covariate factors and significantly outperforms the existing state-of-the-art.

arXiv:1609.06773

Joint CTC-Attention based End-to-End Speech Recognition using Multi-task Learning

Recently, there has been an increasing interest in end-to-end speech recognition that directly transcribes speech to text without any predefined alignments. One approach is the attention-based encoder-decoder framework that learns a mapping between variable-length input and output sequences in one step using a purely data-driven method. The attention model has often been shown to improve the performance over another end-to-end approach, the Connectionist Temporal Classification (CTC), mainly because it explicitly uses the history of the target character without any conditional independence assumptions. However, we observed that the performance of the attention has shown poor results in noisy condition and is hard to learn in the initial training stage with long input sequences. This is because the attention model is too flexible to predict proper alignments in such cases due to the lack of left-to-right constraints as used in CTC. This paper presents a novel method for end-to-end speech recognition to improve robustness and achieve fast convergence by using a joint CTC-attention model within the multi-task learning framework, thereby mitigating the alignment issue. An experiment on the WSJ and CHiME-4 tasks demonstrates its advantages over both the CTC and attention-based encoder-decoder baselines, showing 5.4-14.6% relative improvements in Character Error Rate (CER).

arXiv:1702.01287

Doubly-Attentive Decoder for Multi-modal Neural Machine Translation

We introduce a Multi-modal Neural Machine Translation model in which a doubly-attentive decoder naturally incorporates spatial visual features obtained using pre-trained convolutional neural networks, bridging the gap between image description and translation. Our decoder learns to attend to source-language words and parts of an image independently by means of two separate attention mechanisms as it generates words in the target language. We find that our model can efficiently exploit not just back-translated in-domain multi-modal data but also large general-domain text-only MT corpora. We also report state-of-the-art results on the Multi30k data set.

arXiv:1701.06549

Learning to Decode for Future Success

We introduce a simple, general strategy to manipulate the behavior of a neural decoder that enables it to generate outputs that have specific properties of interest (e.g., sequences of a pre-specified length). The model can be thought of as a simple version of the actor-critic model that uses an interpolation of the actor (the MLE-based token generation policy) and the critic (a value function that estimates the future values of the desired property) for decision making. We demonstrate that the approach is able to incorporate a variety of properties that cannot be handled by standard neural sequence decoders, such as sequence length and backward probability (probability of sources given targets), in addition to yielding consistent improvements in abstractive summarization and machine translation when the property to be optimized is BLEU or ROUGE scores.

arXiv:1702.02429

Trainable Greedy Decoding for Neural Machine Translation

Recent research in neural machine translation has largely focused on two aspects; neural network architectures and end-to-end learning algorithms. The problem of decoding, however, has received relatively little attention from the research community. In this paper, we solely focus on the problem of decoding given a trained neural machine translation model. Instead of trying to build a new decoding algorithm for any specific decoding objective, we propose the idea of trainable decoding algorithm in which we train a decoding algorithm to find a translation that maximizes an arbitrary decoding objective. More specifically, we design an actor that observes and manipulates the hidden state of the neural machine translation decoder and propose to train it using a variant of deterministic policy gradient. We extensively evaluate the proposed algorithm using four language pairs and two decoding objectives and show that we can indeed train a trainable greedy decoder that generates a better translation (in terms of a target decoding objective) with minimal computational overhead.

arXiv:1702.02390공식 코드 · Papers with Code

A Hybrid Convolutional Variational Autoencoder for Text Generation

In this paper we explore the effect of architectural choices on learning a Variational Autoencoder (VAE) for text generation. In contrast to the previously introduced VAE model for text where both the encoder and decoder are RNNs, we propose a novel hybrid architecture that blends fully feed-forward convolutional and deconvolutional components with a recurrent language model. Our architecture exhibits several attractive properties such as faster run time and convergence, ability to better handle long sequences and, more importantly, it helps to avoid some of the major difficulties posed by training VAE models on textual data.