Neural dialogue models, despite their successes, still suffer from lack of relevance, diversity, and in many cases coherence in their generated responses. These issues have been attributed to reasons including (1) short-range model architectures that capture limited temporal dependencies, (2) limitations of the maximum likelihood training objective, (3) the concave entropy profile of dialogue datasets resulting into short and generic responses, and (4) out-of-vocabulary problem leading to generation of a large number of $<$UNK$>$ tokens. Autoregressive transformer models such as GPT-2, although trained with the maximum likelihood objective, do not suffer from the out-of-vocabulary problem and have demonstrated an excellent ability to capture long-range structures in language modeling tasks. In this paper, we examine the use of autoregressive transformer models for multi-turn dialogue response generation. In our experiments, we employ small and medium GPT-2 models (with publicly available pretrained language model parameters) on the open-domain Movie Triples dataset and the closed-domain Ubuntu Dialogue dataset. The models (with and without pretraining) achieve significant improvements over the baselines for multi-turn dialogue response generation. They also produce state-of-the-art performance on the two datasets based on several metrics, including BLEU, ROGUE, and distinct n-gram.

Due to the increased availability of online reviews, sentiment analysis had been witnessed a booming interest from the researchers. Sentiment analysis is a computational treatment of sentiment used to extract and understand the opinions of authors. While many systems were built to predict the sentiment of a document or a sentence, many others provide the necessary detail on various aspects of the entity (i.e. aspect-based sentiment analysis). Most of the available data resources were tailored to English and the other popular European languages. Although Persian is a language with more than 110 million speakers, to the best of our knowledge, there is not any public dataset on aspect-based sentiment analysis in Persian. This paper provides a manually annotated Persian dataset, Pars-ABSA, which is verified by 3 native Persian speakers. The dataset consists of 5114 positive, 3061 negative and 1827 neutral data samples from 5602 unique reviews. Moreover, as a baseline, this paper reports the performance of some state-of-the-art aspect-based sentiment analysis methods with a focus on deep learning, on Pars-ABSA. The obtained results are impressive compared to similar English state-of-the-art.

--Weakly supervised instance labeling using only image-level labels, in lieu of expensive fine-grained pixel annotations, is crucial in several applications including medical image analysis. In contrast to conventional instance segmentation scenarios in computer vision, the problems that we consider are characterized by a small number of training images and non-local patterns that lead to the diagnosis. In this paper, we explore the use of multiple instance learning (MIL) to design an instance label generator under this weakly supervised setting. Motivated by the observation that an MIL model can handle bags of varying sizes, we propose to repurpose an MIL model originally trained for bag-level classification to produce reliable predictions for single instances, i.e., bags of size 1 . T o this end, we introduce a novel regularization strategy based on virtual adversarial training for improving MIL training, and subsequently develop a knowledge distillation technique for repurposing the trained MIL model. Using empirical studies on colon cancer and breast cancer detection from histopathological images, we show that the proposed approach produces high-quality instance-level prediction and significantly outperforms state-of-the MIL methods. For example, in object recognition [1], image-level labels are sufficient to identify generalizable patterns for each category, whereas in instance segmentation [2], [3], pixel-level annotations are required to produce detailed segmentation masks.

This paper proposes a novel method for factorising the information in the latent space of an autoencoder (AE), to improve the interpretability of the latent space and facilitate controlled generation. When trained on a dataset with labelled attributes we can produce a latent vector which separates information encoding the attributes from other characteristic information, and also disentangles the attribute information. This then allows us to manipulate each attribute of the latent representation individually without affecting others. Our method, matrix subspace projection, is simpler than the state of the art adversarial network approaches to latent space factorisation. We demonstrate the utility of the method for attribute manipulation tasks on the CelebA image dataset and the E2E text corpus.

Model compression has emerged as an important area of research for deploying deep learning models on Internet-of-Things (IoT). However, for extremely memory-constrained scenarios, even the compressed models cannot fit within the memory of a single device and, as a result, must be distributed across multiple devices. This leads to a distributed inference paradigm in which memory and communication costs represent a major bottleneck. Yet, existing model compression techniques are not communication-aware. Therefore, we propose Network of Neural Networks (NoNN), a new distributed IoT learning paradigm that compresses a large pretrained 'teacher' deep network into several disjoint and highly-compressed 'student' modules, without loss of accuracy. Moreover, we propose a network science-based knowledge partitioning algorithm for the teacher model, and then train individual students on the resulting disjoint partitions. Extensive experimentation on five image classification datasets, for user-defined memory/performance budgets, show that NoNN achieves higher accuracy than several baselines and similar accuracy as the teacher model, while using minimal communication among students. Finally, as a case study, we deploy the proposed model for CIFAR-10 dataset on edge devices and demonstrate significant improvements in memory footprint (up to 24x), performance (up to 12x), and energy per node (up to 14x) compared to the large teacher model. We further show that for distributed inference on multiple edge devices, our proposed NoNN model results in up to 33x reduction in total latency w.r.t. a state-of-the-art model compression baseline.

We present a novel unsupervised deep learning approach that utilizes the encoder-decoder architecture for detecting anomalies in sequential sensor data collected during industrial manufacturing. Our approach is designed not only to detect whether there exists an anomaly at a given time step, but also to predict what will happen next in the (sequential) process. We demonstrate our approach on a dataset collected from a real-world testbed. The dataset contains images collected under both normal conditions and synthetic anomalies. We show that the encoder-decoder model is able to identify the injected anomalies in a modern manufacturing process in an unsupervised fashion. In addition, it also gives hints about the temperature non-uniformity of the testbed during manufacturing, which is what we are not aware of before doing the experiment.

ABSTRACT In recent years, search story, a combined display with other organic channels, has become a major source of user traffic on platforms such as e-commerce search platforms, news feed platforms and web and image search platforms. The recommended search story guides a user to identify her own preference and personal intent, which subsequently influences the user's real-time and long-term search behavior. As search stories become increasingly important, in this work, we study the problem of personalized search story recommendation within a search engine, which aims to suggest a search story relevant to both a search keyword and an individual user's interest. To address the challenge of modeling both immediate and future values of recommended search stories (i.e., cross-channel effect), for which conventional supervised learning framework is not applicable, we resort to a Markov decision process and propose a deep reinforcement learning architecture trained by both imitation learning and reinforcement learning. We empirically demonstrate the effectiveness of our proposed approach through extensive experiments on real-world data sets from JD.com. 1. INTRODUCTION Imagine that a customer visits a retail shop to purchase a dress which is to her liking. As the customer walks in, a business assistant is present to assist the customer by answering questions on fashion trend or suggesting related dresses. In online e-commerce applications, more business units are adding a component that plays a similar role as the business assistant in a shop. In this paper, we are interested in a particular component, commonly known as search story, that has become popular among e-commerce search engines on many online platforms. For instance, in news feed platforms and web and image search platforms, each search story is a display of recommended high-quality content which is relevant to a user's personal interests. In e-commerce search (a) Display search story within organic product item search page (b) Landing page after clicking search story, which contains both shopping guides and shopping product items Figure 1: An illustrated (not a screenshot) example of search story recommendation.

Capsules are the multidimensional analogue to scalar neurons in neural networks, and because they are multidimensional, much more complex routing schemes can be used to pass information forward through the network than what can be used in traditional neural networks. This work treats capsules as collections of neurons in a fully connected neural network, where sub-networks connecting capsules are weighted according to the routing coefficients determined by routing by agreement. An energy function is designed to reflect this model, and it follows that capsule networks with dynamic routing can be formulated as a product of expert neurons. By alternating between dynamic routing, which acts to both find subnetworks within the overall network as well as to mix the model distribution, and updating the parameters by the gradient of the contrastive divergence, a bottom-up, unsupervised learning algorithm is constructed for capsule networks with dynamic routing. The model and its training algorithm are qualitatively tested in the generative sense, and is able to produce realistic looking images from standard vision datasets.

In this paper, we introduce multi-task learning (MTL) to data harmonization (DH); where we aim to harmonize images across different acquisition platforms and sites. This allows us to integrate information from multiple acquisitions and improve the predictive performance and learning efficiency of the harmonization model. Specifically, we introduce the Multi Stage Prediction (MSP) Network, a MTL framework that incorporates neural networks of potentially disparate architectures, trained for different individual acquisition platforms, into a larger architecture that is refined in unison. The MSP utilizes high-level features of single networks for individual tasks, as inputs of additional neural networks to inform the final prediction, therefore exploiting redundancy across tasks to make the most of limited training data. We validate our methods on a dMRI harmonization challenge dataset, where we predict three modern platform types, from one obtained from an old scanner. We show how MTL architectures, such as the MSP, produce around 20\% improvement of patch-based mean-squared error over current state-of-the-art methods and that our MSP outperforms off-the-shelf MTL networks. Our code is available https://github.com/sbb-gh/ .

Although distributed computing can significantly reduce the training time of deep neural networks, scaling the training process while maintaining high efficiency and final accuracy is challenging. Distributed asynchronous training enjoys near-linear speedup, but asynchrony causes gradient staleness, the main difficulty in scaling stochastic gradient descent to large clusters. Momentum, which is often used to accelerate convergence and escape local minima, exacerbates the gradient staleness, thereby hindering convergence. We propose DANA: a novel asynchronous distributed technique which is based on a new gradient staleness measure that we call the gap. By minimizing the gap, DANA mitigates the gradient staleness, despite using momentum, and therefore scales to large clusters while maintaining high final accuracy and fast convergence. DANA adapts Nesterov's Accelerated Gradient to a distributed setting, computing the gradient on an estimated future position of the model's parameters. In turn, we show that DANA's estimation of the future position amplifies the use of a Taylor expansion, which relies on a fast Hessian approximation, making it much more effective and accurate. Our evaluation on the CIFAR and ImageNet datasets shows that DANA outperforms existing methods, in both final accuracy and convergence speed.