Fork me on GitHub

Trending arXiv

Note: this version is tailored to @Smerity - though you can run your own! Trending arXiv may eventually be extended to multiple users ...


1 2 3 4 5 6 7 8 9 35 36

Close to Human Quality TTS with Transformer

Naihan Li, Shujie Liu, Yanqing Liu, Sheng Zhao, Ming Liu, Ming Zhou

Although end-to-end neural text-to-speech (TTS) methods (such as Tacotron2) are proposed and achieve state-of-the-art performance, they still suffer from two problems: 1) low efficiency during training and inference; 2) hard to model long dependency using current recurrent neural networks (RNNs). Inspired by the success of Transformer network in neural machine translation (NMT), in this paper, we introduce and adapt the multi-head attention mechanism to replace the RNN structures and also the original attention mechanism in Tacotron2. With the help of multi-head self-attention, the hidden states in the encoder and decoder are constructed in parallel, which improves training efficiency. Meanwhile, any two inputs at different times are connected directly by a self-attention mechanism, which solves the long range dependency problem effectively. Using phoneme sequences as input, our Transformer TTS network generates mel spectrograms, followed by a WaveNet vocoder to output the final audio results. Experiments are conducted to test the efficiency and performance of our new network. For the efficiency, our Transformer TTS network can speed up the training about 4.25 times faster compared with Tacotron2. For the performance, rigorous human tests show that our proposed model achieves state-of-the-art performance (outperforms Tacotron2 with a gap of 0.048) and is very close to human quality (4.39 vs 4.44 in MOS).

Captured tweets and retweets: 2

Addressing the Fundamental Tension of PCGML with Discriminative Learning

Isaac Karth, Adam M. Smith

Procedural content generation via machine learning (PCGML) is typically framed as the task of fitting a generative model to full-scale examples of a desired content distribution. This approach presents a fundamental tension: the more design effort expended to produce detailed training examples for shaping a generator, the lower the return on investment from applying PCGML in the first place. In response, we propose the use of discriminative models (which capture the validity of a design rather the distribution of the content) trained on positive and negative examples. Through a modest modification of WaveFunctionCollapse, a commercially-adopted PCG approach that we characterize as using elementary machine learning, we demonstrate a new mode of control for learning-based generators. We demonstrate how an artist might craft a focused set of additional positive and negative examples by critique of the generator's previous outputs. This interaction mode bridges PCGML with mixed-initiative design assistance tools by working with a machine to define a space of valid designs rather than just one new design.

Captured tweets and retweets: 2

Interpretable Visual Question Answering by Reasoning on Dependency Trees

Qingxing Cao, Xiaodan Liang, Bailin Li, Liang Lin

Collaborative reasoning for understanding each image-question pair is very critical but underexplored for an interpretable visual question answering system. Although very recent works also attempted to use explicit compositional processes to assemble multiple subtasks embedded in the questions, their models heavily rely on annotations or handcrafted rules to obtain valid reasoning processes, leading to either heavy workloads or poor performance on composition reasoning. In this paper, to better align image and language domains in diverse and unrestricted cases, we propose a novel neural network model that performs global reasoning on a dependency tree parsed from the question, and we thus phrase our model as parse-tree-guided reasoning network (PTGRN). This network consists of three collaborative modules: i) an attention module to exploit the local visual evidence for each word parsed from the question, ii) a gated residual composition module to compose the previously mined evidence, and iii) a parse-tree-guided propagation module to pass the mined evidence along the parse tree. Our PTGRN is thus capable of building an interpretable VQA system that gradually derives the image cues following a question-driven parse-tree reasoning route. Experiments on relational datasets demonstrate the superiority of our PTGRN over current state-of-the-art VQA methods, and the visualization results highlight the explainable capability of our reasoning system.

Captured tweets and retweets: 2

Appendix - Recommended Statistical Significance Tests for NLP Tasks

Rotem Dror, Roi Reichart

Statistical significance testing plays an important role when drawing conclusions from experimental results in NLP papers. Particularly, it is a valuable tool when one would like to establish the superiority of one algorithm over another. This appendix complements the guide for testing statistical significance in NLP presented in \cite{dror2018hitchhiker} by proposing valid statistical tests for the common tasks and evaluation measures in the field.

Captured tweets and retweets: 2

Recurrent World Models Facilitate Policy Evolution

David Ha, Jürgen Schmidhuber

A generative recurrent neural network is quickly trained in an unsupervised manner to model popular reinforcement learning environments through compressed spatio-temporal representations. The world model's extracted features are fed into compact and simple policies trained by evolution, achieving state of the art results in various environments. We also train our agent entirely inside of an environment generated by its own internal world model, and transfer this policy back into the actual environment. Interactive version of paper at

Captured tweets and retweets: 1

When to Finish? Optimal Beam Search for Neural Text Generation (modulo beam size)

Liang Huang, Kai Zhao, Mingbo Ma

In neural text generation such as neural machine translation, summarization, and image captioning, beam search is widely used to improve the output text quality. However, in the neural generation setting, hypotheses can finish in different steps, which makes it difficult to decide when to end beam search to ensure optimality. We propose a provably optimal beam search algorithm that will always return the optimal-score complete hypothesis (modulo beam size), and finish as soon as the optimality is established (finishing no later than the baseline). To counter neural generation's tendency for shorter hypotheses, we also introduce a bounded length reward mechanism which allows a modified version of our beam search algorithm to remain optimal. Experiments on neural machine translation demonstrate that our principled beam search algorithm leads to improvement in BLEU score over previously proposed alternatives.

Captured tweets and retweets: 2

Spherical Latent Spaces for Stable Variational Autoencoders

Jiacheng Xu, Greg Durrett

A hallmark of variational autoencoders (VAEs) for text processing is their combination of powerful encoder-decoder models, such as LSTMs, with simple latent distributions, typically multivariate Gaussians. These models pose a difficult optimization problem: there is an especially bad local optimum where the variational posterior always equals the prior and the model does not use the latent variable at all, a kind of "collapse" which is encouraged by the KL divergence term of the objective. In this work, we experiment with another choice of latent distribution, namely the von Mises-Fisher (vMF) distribution, which places mass on the surface of the unit hypersphere. With this choice of prior and posterior, the KL divergence term now only depends on the variance of the vMF distribution, giving us the ability to treat it as a fixed hyperparameter. We show that doing so not only averts the KL collapse, but consistently gives better likelihoods than Gaussians across a range of modeling conditions, including recurrent language modeling and bag-of-words document modeling. An analysis of the properties of our vMF representations shows that they learn richer and more nuanced structures in their latent representations than their Gaussian counterparts.

Captured tweets and retweets: 1

FPGA Implementation of Convolutional Neural Networks with Fixed-Point Calculations

Roman A. Solovyev, Alexandr A. Kalinin, Alexander G. Kustov, Dmitry V. Telpukhov, Vladimir S. Ruhlov

Neural network-based methods for image processing are becoming widely used in practical applications. Modern neural networks are computationally expensive and require specialized hardware, such as graphics processing units. Since such hardware is not always available in real life applications, there is a compelling need for the design of neural networks for mobile devices. Mobile neural networks typically have reduced number of parameters and require a relatively small number of arithmetic operations. However, they usually still are executed at the software level and use floating-point calculations. The use of mobile networks without further optimization may not provide sufficient performance when high processing speed is required, for example, in real-time video processing (30 frames per second). In this study, we suggest optimizations to speed up computations in order to efficiently use already trained neural networks on a mobile device. Specifically, we propose an approach for speeding up neural networks by moving computation from software to hardware and by using fixed-point calculations instead of floating-point. We propose a number of methods for neural network architecture design to improve the performance with fixed-point calculations. We also show an example of how existing datasets can be modified and adapted for the recognition task in hand. Finally, we present the design and the implementation of a floating-point gate array-based device to solve the practical problem of real-time handwritten digit classification from mobile camera video feed.

Captured tweets and retweets: 1

Approximate Exploration through State Abstraction

Adrien Ali Taïga, Aaron Courville, Marc G. Bellemare

Although exploration in reinforcement learning is well understood from a theoretical point of view, provably correct methods remain impractical. In this paper we study the interplay between exploration and approximation, what we call \emph{approximate exploration}. We first provide results when the approximation is explicit, quantifying the performance of an exploration algorithm, MBIE-EB \citep{strehl2008analysis}, when combined with state aggregation. In particular, we show that this allows the agent to trade off between learning speed and quality of the policy learned. We then turn to a successful exploration scheme in practical, pseudo-count based exploration bonuses \citep{bellemare2016unifying}. We show that choosing a density model implicitly defines an abstraction and that the pseudo-count bonus incentivizes the agent to explore using this abstraction. We find, however, that implicit exploration may result in a mismatch between the approximated value function and exploration bonus, leading to either under- or over-exploration.

Captured tweets and retweets: 2

End-to-End Neural Entity Linking

Nikolaos Kolitsas, Octavian-Eugen Ganea, Thomas Hofmann

Entity Linking (EL) is an essential task for semantic text understanding and information extraction. Popular methods separately address the Mention Detection (MD) and Entity Disambiguation (ED) stages of EL, without leveraging their mutual dependency. We here propose the first neural end-to-end EL system that jointly discovers and links entities in a text document. The main idea is to consider all possible spans as potential mentions and learn contextual similarity scores over their entity candidates that are useful for both MD and ED decisions. Key components are context-aware mention embeddings, entity embeddings and a probabilistic mention - entity map, without demanding other engineered features. Empirically, we show that our end-to-end method significantly outperforms popular systems on the Gerbil platform when enough training data is available. Conversely, if testing datasets follow different annotation conventions compared to the training set (e.g. queries/ tweets vs news documents), our ED model coupled with a traditional NER system offers the best or second best EL accuracy.

Captured tweets and retweets: 2

Increasing Trust in AI Services through Supplier's Declarations of Conformity

Michael Hind, Sameep Mehta, Aleksandra Mojsilovic, Ravi Nair, Karthikeyan Natesan Ramamurthy, Alexandra Olteanu, Kush R. Varshney

The accuracy and reliability of machine learning algorithms are an important concern for suppliers of artificial intelligence (AI) services, but considerations beyond accuracy, such as safety, security, and provenance, are also critical elements to engender consumers' trust in a service. In this paper, we propose a supplier's declaration of conformity (SDoC) for AI services to help increase trust in AI services. An SDoC is a transparent, standardized, but often not legally required, document used in many industries and sectors to describe the lineage of a product along with the safety and performance testing it has undergone. We envision an SDoC for AI services to contain purpose, performance, safety, security, and provenance information to be completed and voluntarily released by AI service providers for examination by consumers. Importantly, it conveys product-level rather than component-level functional testing. We suggest a set of declaration items tailored to AI and provide examples for two fictitious AI services.

Captured tweets and retweets: 1

An elementary introduction to information geometry

Frank Nielsen

We describe the fundamental differential-geometric structures of information manifolds, state the fundamental theorem of information geometry, and illustrate some uses of these information manifolds in information sciences. The exposition is self-contained by concisely introducing the necessary concepts of differential geometry with proofs omitted for brevity.

Captured tweets and retweets: 2

Learning Invariances using the Marginal Likelihood

Mark van der Wilk, Matthias Bauer, ST John, James Hensman

Generalising well in supervised learning tasks relies on correctly extrapolating the training data to a large region of the input space. One way to achieve this is to constrain the predictions to be invariant to transformations on the input that are known to be irrelevant (e.g. translation). Commonly, this is done through data augmentation, where the training set is enlarged by applying hand-crafted transformations to the inputs. We argue that invariances should instead be incorporated in the model structure, and learned using the marginal likelihood, which correctly rewards the reduced complexity of invariant models. We demonstrate this for Gaussian process models, due to the ease with which their marginal likelihood can be estimated. Our main contribution is a variational inference scheme for Gaussian processes containing invariances described by a sampling procedure. We learn the sampling procedure by back-propagating through it to maximise the marginal likelihood.

Captured tweets and retweets: 2

Deep RTS: A Game Environment for Deep Reinforcement Learning in Real-Time Strategy Games

Per-Arne Andersen, Morten Goodwin, Ole-Christoffer Granmo

Reinforcement learning (RL) is an area of research that has blossomed tremendously in recent years and has shown remarkable potential for artificial intelligence based opponents in computer games. This success is primarily due to the vast capabilities of convolutional neural networks, that can extract useful features from noisy and complex data. Games are excellent tools to test and push the boundaries of novel RL algorithms because they give valuable insight into how well an algorithm can perform in isolated environments without the real-life consequences. Real-time strategy games (RTS) is a genre that has tremendous complexity and challenges the player in short and long-term planning. There is much research that focuses on applied RL in RTS games, and novel advances are therefore anticipated in the not too distant future. However, there are to date few environments for testing RTS AIs. Environments in the literature are often either overly simplistic, such as microRTS, or complex and without the possibility for accelerated learning on consumer hardware like StarCraft II. This paper introduces the Deep RTS game environment for testing cutting-edge artificial intelligence algorithms for RTS games. Deep RTS is a high-performance RTS game made specifically for artificial intelligence research. It supports accelerated learning, meaning that it can learn at a magnitude of 50 000 times faster compared to existing RTS games. Deep RTS has a flexible configuration, enabling research in several different RTS scenarios, including partially observable state-spaces and map complexity. We show that Deep RTS lives up to our promises by comparing its performance with microRTS, ELF, and StarCraft II on high-end consumer hardware. Using Deep RTS, we show that a Deep Q-Network agent beats random-play agents over 70% of the time. Deep RTS is publicly available at

Captured tweets and retweets: 2

Top-Down Tree Structured Text Generation

Qipeng Guo, Xipeng Qiu, Xiangyang Xue, Zheng Zhang

Text generation is a fundamental building block in natural language processing tasks. Existing sequential models performs autoregression directly over the text sequence and have difficulty generating long sentences of complex structures. This paper advocates a simple approach that treats sentence generation as a tree-generation task. By explicitly modelling syntactic structures in a constituent syntactic tree and performing top-down, breadth-first tree generation, our model fixes dependencies appropriately and performs implicit global planning. This is in contrast to transition-based depth-first generation process, which has difficulty dealing with incomplete texts when parsing and also does not incorporate future contexts in planning. Our preliminary results on two generation tasks and one parsing task demonstrate that this is an effective strategy.

Captured tweets and retweets: 2

Character-Level Language Modeling with Deeper Self-Attention

Rami Al-Rfou, Dokook Choe, Noah Constant, Mandy Guo, Llion Jones

LSTMs and other RNN variants have shown strong performance on character-level language modeling. These models are typically trained using truncated backpropagation through time, and it is common to assume that their success stems from their ability to remember long-term contexts. In this paper, we show that a deep (64-layer) transformer model with fixed context outperforms RNN variants by a large margin, achieving state of the art on two popular benchmarks- 1.13 bits per character on text8 and 1.06 on enwik8. To get good results at this depth, we show that it is important to add auxiliary losses, both at intermediate network layers and intermediate sequence positions.

Captured tweets and retweets: 2

Predicting Expressive Speaking Style From Text In End-To-End Speech Synthesis

Daisy Stanton, Yuxuan Wang, RJ Skerry-Ryan

Global Style Tokens (GSTs) are a recently-proposed method to learn latent disentangled representations of high-dimensional data. GSTs can be used within Tacotron, a state-of-the-art end-to-end text-to-speech synthesis system, to uncover expressive factors of variation in speaking style. In this work, we introduce the Text-Predicted Global Style Token (TP-GST) architecture, which treats GST combination weights or style embeddings as "virtual" speaking style labels within Tacotron. TP-GST learns to predict stylistic renderings from text alone, requiring neither explicit labels during training nor auxiliary inputs for inference. We show that, when trained on a dataset of expressive speech, our system generates audio with more pitch and energy variation than two state-of-the-art baseline models. We further demonstrate that TP-GSTs can synthesize speech with background noise removed, and corroborate these analyses with positive results on human-rated listener preference audiobook tasks. Finally, we demonstrate that multi-speaker TP-GST models successfully factorize speaker identity and speaking style. We provide a website with audio samples for each of our findings.

Captured tweets and retweets: 2

Diverse Image-to-Image Translation via Disentangled Representations

Hsin-Ying Lee, Hung-Yu Tseng, Jia-Bin Huang, Maneesh Kumar Singh, Ming-Hsuan Yang

Image-to-image translation aims to learn the mapping between two visual domains. There are two main challenges for many applications: 1) the lack of aligned training pairs and 2) multiple possible outputs from a single input image. In this work, we present an approach based on disentangled representation for producing diverse outputs without paired training images. To achieve diversity, we propose to embed images onto two spaces: a domain-invariant content space capturing shared information across domains and a domain-specific attribute space. Our model takes the encoded content features extracted from a given input and the attribute vectors sampled from the attribute space to produce diverse outputs at test time. To handle unpaired training data, we introduce a novel cross-cycle consistency loss based on disentangled representations. Qualitative results show that our model can generate diverse and realistic images on a wide range of tasks without paired training data. For quantitative comparisons, we measure realism with user study and diversity with a perceptual distance metric. We apply the proposed model to domain adaptation and show competitive performance when compared to the state-of-the-art on the MNIST-M and the LineMod datasets.

Captured tweets and retweets: 2

Learning Dexterous In-Hand Manipulation

OpenAI, Marcin Andrychowicz, Bowen Baker, Maciek Chociej, Rafal Jozefowicz, Bob McGrew, Jakub Pachocki, Arthur Petron, Matthias Plappert, Glenn Powell, Alex Ray, Jonas Schneider, Szymon Sidor, Josh Tobin, Peter Welinder, Lilian Weng, Wojciech Zaremba

We use reinforcement learning (RL) to learn dexterous in-hand manipulation policies which can perform vision-based object reorientation on a physical Shadow Dexterous Hand. The training is performed in a simulated environment in which we randomize many of the physical properties of the system like friction coefficients and an object's appearance. Our policies transfer to the physical robot despite being trained entirely in simulation. Our method does not rely on any human demonstrations, but many behaviors found in human manipulation emerge naturally, including finger gaiting, multi-finger coordination, and the controlled use of gravity. Our results were obtained using the same distributed RL system that was used to train OpenAI Five. We also include a video of our results:

Captured tweets and retweets: 2

Highly Scalable Deep Learning Training System with Mixed-Precision: Training ImageNet in Four Minutes

Xianyan Jia, Shutao Song, Wei He, Yangzihao Wang, Haidong Rong, Feihu Zhou, Liqiang Xie, Zhenyu Guo, Yuanzhou Yang, Liwei Yu, Tiegang Chen, Guangxiao Hu, Shaohuai Shi, Xiaowen Chu

Synchronized stochastic gradient descent (SGD) optimizers with data parallelism are widely used in training large-scale deep neural networks. Although using larger mini-batch sizes can improve the system scalability by reducing the communication-to-computation ratio, it may hurt the generalization ability of the models. To this end, we build a highly scalable deep learning training system for dense GPU clusters with three main contributions: (1) We propose a mixed-precision training method that significantly improves the training throughput of a single GPU without losing accuracy. (2) We propose an optimization approach for extremely large mini-batch size (up to 64k) that can train CNN models on the ImageNet dataset without losing accuracy. (3) We propose highly optimized all-reduce algorithms that achieve up to 3x and 11x speedup on AlexNet and ResNet-50 respectively than NCCL-based training on a cluster with 1024 Tesla P40 GPUs. On training ResNet-50 with 90 epochs, the state-of-the-art GPU-based system with 1024 Tesla P100 GPUs spent 15 minutes and achieved 74.9\% top-1 test accuracy, and another KNL-based system with 2048 Intel KNLs spent 20 minutes and achieved 75.4\% accuracy. Our training system can achieve 75.8\% top-1 test accuracy in only 6.6 minutes using 2048 Tesla P40 GPUs. When training AlexNet with 95 epochs, our system can achieve 58.7\% top-1 test accuracy within 4 minutes, which also outperforms all other existing systems.

Captured tweets and retweets: 63

1 2 3 4 5 6 7 8 9 35 36