Multi-Component Optimization and Efficient Deployment of Neural-Networks on Resource-Constrained IoT Hardware

The majority of IoT devices like smartwatches, smart plugs, HVAC controllers, etc., are powered by hardware with a constrained specification (low memory, clock speed and processor) which is insufficient to accommodate and execute large, high-quality models. On such resource-constrained devices, manufacturers still manage to provide attractive functionalities (to boost sales) by following the traditional approach of programming IoT devices/products to collect and transmit data (image, audio, sensor readings, etc.) to their cloud-based ML analytics platforms. For decades, this online approach has been facing issues such as compromised data streams, non-real-time analytics due to latency, bandwidth constraints, costly subscriptions, recent privacy issues raised by users and the GDPR guidelines, etc. In this paper, to enable ultra-fast and accurate AI-based offline analytics on resource-constrained IoT devices, we present an end-to-end multi-component model optimization sequence and open-source its implementation. Researchers and developers can use our optimization sequence to optimize high memory, computation demanding models in multiple aspects in order to produce small size, low latency, low-power consuming models that can comfortably fit and execute on resource-constrained hardware. The experimental results show that our optimization components can produce models that are; (i) 12.06 x times compressed; (ii) 0.13% to 0.27% more accurate; (iii) Orders of magnitude faster unit inference at 0.06 ms. Our optimization sequence is generic and can be applied to any state-of-the-art models trained for anomaly detection, predictive maintenance, robotics, voice recognition, and machine vision.

Data Selection Strategies for Multi-Domain Sentiment Analysis

Domain adaptation is important in sentiment analysis as sentiment-indicating words vary between domains. Recently, multi-domain adaptation has become more pervasive, but existing approaches train on all available source domains including dissimilar ones. However, the selection of appropriate training data is as important as the choice of algorithm. We undertake -- to our knowledge for the first time -- an extensive study of domain similarity metrics in the context of sentiment analysis and propose novel representations, metrics, and a new scope for data selection. We evaluate the proposed methods on two large-scale multi-domain adaptation settings on tweets and reviews and demonstrate that they consistently outperform strong random and balanced baselines, while our proposed selection strategy outperforms instance-level selection and yields the best score on a large reviews corpus.

Knowledge Adaptation: Teaching to Adapt

Domain adaptation is crucial in many real-world applications where the distribution of the training data differs from the distribution of the test data. Previous Deep Learning-based approaches to domain adaptation need to be trained jointly on source and target domain data and are therefore unappealing in scenarios where models need to be adapted to a large number of domains or where a domain is evolving, e.g. spam detection where attackers continuously change their tactics. To fill this gap, we propose Knowledge Adaptation, an extension of Knowledge Distillation (Bucilua et al., 2006; Hinton et al., 2015) to the domain adaptation scenario. We show how a student model achieves state-of-the-art results on unsupervised domain adaptation from multiple sources on a standard sentiment analysis benchmark by taking into account the domain-specific expertise of multiple teachers and the similarities between their domains. When learning from a single teacher, using domain similarity to gauge trustworthiness is inadequate. To this end, we propose a simple metric that correlates well with the teacher's accuracy in the target domain. We demonstrate that incorporating high-confidence examples selected by this metric enables the student model to achieve state-of-the-art performance in the single-source scenario.

Towards a continuous modeling of natural language domains

Humans continuously adapt their style and language to a variety of domains. However, a reliable definition of `domain' has eluded researchers thus far. Additionally, the notion of discrete domains stands in contrast to the multiplicity of heterogeneous domains that humans navigate, many of which overlap. In order to better understand the change and variation of human language, we draw on research in domain adaptation and extend the notion of discrete domains to the continuous spectrum. We propose representation learning-based models that can adapt to continuous domains and detail how these can be used to investigate variation in language. To this end, we propose to use dialogue modeling as a test bed due to its proximity to language modeling and its social component.

INSIGHT-1 at SemEval-2016 Task 5: Deep Learning for Multilingual Aspect-based Sentiment Analysis

This paper describes our deep learning-based approach to multilingual aspect-based sentiment analysis as part of SemEval 2016 Task 5. We use a convolutional neural network (CNN) for both aspect extraction and aspect-based sentiment analysis. We cast aspect extraction as a multi-label classification problem, outputting probabilities over aspects parameterized by a threshold. To determine the sentiment towards an aspect, we concatenate an aspect vector with every word embedding and apply a convolution over it. Our constrained system (unconstrained for English) achieves competitive results across all languages and domains, placing first or second in 5 and 7 out of 11 language-domain pairs for aspect category detection (slot 1) and sentiment polarity (slot 3) respectively, thereby demonstrating the viability of a deep learning-based approach for multilingual aspect-based sentiment analysis.

Character-level and Multi-channel Convolutional Neural Networks for Large-scale Authorship Attribution

Convolutional neural networks (CNNs) have demonstrated superior capability for extracting information from raw signals in computer vision. Recently, character-level and multi-channel CNNs have exhibited excellent performance for sentence classification tasks. We apply CNNs to large-scale authorship attribution, which aims to determine an unknown text's author among many candidate authors, motivated by their ability to process character-level signals and to differentiate between a large number of classes, while making fast predictions in comparison to state-of-the-art approaches. We extensively evaluate CNN-based approaches that leverage word and character channels and compare them against state-of-the-art methods for a large range of author numbers, shedding new light on traditional approaches. We show that character-level CNNs outperform the state-of-the-art on four out of five datasets in different domains. Additionally, we present the first application of authorship attribution to reddit.

INSIGHT-1 at SemEval-2016 Task 4: Convolutional Neural Networks for Sentiment Classification and Quantification

This paper describes our deep learning-based approach to sentiment analysis in Twitter as part of SemEval-2016 Task 4. We use a convolutional neural network to determine sentiment and participate in all subtasks, i.e. two-point, three-point, and five-point scale sentiment classification and two-point and five-point scale sentiment quantification. We achieve competitive results for two-point scale sentiment classification and quantification, ranking fifth and a close fourth (third and second by alternative metrics) respectively despite using only pre-trained embeddings that contain no sentiment information. We achieve good performance on three-point scale sentiment classification, ranking eighth out of 35, while performing poorly on five-point scale sentiment classification and quantification. An error analysis reveals that this is due to low expressiveness of the model to capture negative sentiment as well as an inability to take into account ordinal information. We propose improvements in order to address these and other issues.

A Hierarchical Model of Reviews for Aspect-based Sentiment Analysis

Opinion mining from customer reviews has become pervasive in recent years. Sentences in reviews, however, are usually classified independently, even though they form part of a review's argumentative structure. Intuitively, sentences in a review build and elaborate upon each other; knowledge of the review structure and sentential context should thus inform the classification of each sentence. We demonstrate this hypothesis for the task of aspect-based sentiment analysis by modeling the interdependencies of sentences in a review with a hierarchical bidirectional LSTM. We show that the hierarchical model outperforms two non-hierarchical baselines, obtains results competitive with the state-of-the-art, and outperforms the state-of-the-art on five multilingual, multi-domain datasets without any hand-engineered features or external resources.