Papers

Wind energy is an important source of renewable and sustainable energy and therefore an elementary component of any future energy supply. However, the operation of large wind farms places high demands on reliability and is often impacted by high maintenance and repair costs in the event of a failure. A frequency converter is one of the most important components of each wind turbine, which ensures that the frequency of the generated energy synchronises with the grid frequency and thus enables the flow of energy into the power grid. The detection of anomalies in these devices is complex due to the high frequency and multidimensionality of different sensor information from the energy control units and requires fault patterns to be discovered and detected in large time series. In this paper, we show how state-of-the-art self-supervised-learning techniques, namely LSTM autoencoders, can be successfully applied to real-world data. We describe the extensions we have made to deal with the often very noisy sensors and describe the construction of the training data set. The trained system was first tested and evaluated on synthetic data and subsequently on a large real-world data set. In both cases, it was shown that outliers can be reliably identified using our presented approach.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7

Digitization of newspapers is of interest for many reasons including preservation of history, accessibility and search ability, etc. While digitization of documents such as scientific articles and magazines is prevalent in literature, one of the main challenges for digitization of newspaper lies in its complex layout (e.g. articles spanning multiple columns, text interrupted by images) analysis, which is necessary to preserve human read-order. This work provides a major breakthrough in the digitization of newspapers on three fronts: first, releasing a dataset of 3000 fully-annotated, real-world newspaper images from 21 different U.S. states representing an extensive variety of complex layouts for document layout analysis; second, proposing layout segmentation as a precursor to existing optical character recognition (OCR) engines, where multiple state-of-the-art image segmentation models and several post-processing methods are explored for document layout segmentation; third, providing a thorough and structured evaluation protocol for isolated layout segmentation and end-to-end OCR.

poster session 7 @ red 6, poster session 8 @ red 6, poster session 7, poster session 8

We propose PaintTeR, our Paintings TextRank algorithm for extracting art-related text spans from passages on paintings. PaintTeR combines a lexicon of painting words curated automatically through distant supervision with random walks on a large-scale word co-occurrence graph for ranking passage spans for artistic characteristics. The spans extracted with PaintTeR are used in state-of-the-art Question Generation and Reading Comprehension models for designing an interactive aid that enables gallery and museum visitors focus on the artistic elements of paintings. We provide experiments on two datasets of expert-written passages on paintings to showcase the effectiveness of PaintTeR. Evaluations by both gallery experts as well as crowdworkers indicate that our proposed algorithm can be used to select relevant and interesting art-centered questions. To the best of our knowledge, ours is the first work to effectively fine-tune question generation models using minimal supervision for a low-resource, specialized context such as gallery visits.

poster session 7 @ red 6, poster session 8 @ red 6, poster session 7, poster session 8

Predictive maintenance systems have the potential to significantly reduce costs for maintaining aircraft fleets as well as provide improved safety by detecting maintenance issues before they come severe. However, the development of such systems has been limited due to a lack of publicly labeled multivariate time series (MTS) sensor data. MTS classification has advanced greatly over the past decade, but there is a lack of sufficiently challenging benchmarks for new methods. This work introduces the NGAFID Maintenance Classification (NGAFID-MC) dataset as a novel benchmark in terms of difficulty, number of samples, and sequence length. NGAFID-MC consists of over 7,500 labeled flights, representing over 11,500 hours of per second flight data recorder readings of 23 sensor parameters. Using this benchmark, we demonstrate that Recurrent Neural Network (RNN) methods are not well suited for capturing temporally distant relationships and propose a new architecture called Convolutional Multiheaded Self Attention (Conv-MHSA) that achieves greater classification performance at greater computational efficiency. We also demonstrate that image inspired augmentations of cutout, mixup, and cutmix, can be used to reduce overfitting and improve generalization in MTS classification. Our best trained models have been incorporated back into the NGAFID to allow users to potentially detect flights that require maintenance as well as provide feedback to further expand and refine the NGAFID-MC dataset.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7

Recent years have seen significant advances in multi-turn Spoken Language Understanding (SLU), where dialogue contexts are used to guide intent classification and slot filling. However, how to selectively incorporate dialogue contexts, such as previous utterances and dialogue acts, into multi-turn SLU still remains a substantial challenge. In this work, we propose a novel contextual SLU model for multi-turn intent classification and slot filling tasks. We introduce an adaptive global-local context fusion mechanism to selectively integrate dialogue contexts into our model. The local context fusion aligns each dialogue context using multi-head attention, while the global context fusion measures overall context contribution to intent classification and slot filling tasks. Experiments show that on two benchmark datasets, our model achieves absolute F1 score improvements of 2.73% and 2.57% for the slot filling task on Sim-R and Sim M datasets, respectively. Additional experiments on a large-scale, de-identified, in-house dataset further verify the measurable accuracy gains of our proposed model.

poster session 7 @ red 6, poster session 8 @ red 6, poster session 7, poster session 8

Fuzzing is a widely-used testing technique to assure software robustness. However, automatic generation of high-quality test suites is challenging, especially for software that takes in highly-structured inputs, such as the compilers. Compiler fuzzing remains difficult as generating tons of syntactically and semantically valid programs is not trivial. Most previous methods either depend on human-crafted grammars or heuristics to learn partial language patterns. They both suffer from the completeness issue that is a classic puzzle in software testing. To mitigate the problem, we propose a knowledge-guided reinforcement learning-based approach to generating valid programs for compiler fuzzing. We first design a naive learning model which evolves with the sequential mutation rewards provided by a target compiler we test. By iterating the training cycle, the model learns to generate valid programs that can improve the testing efficacy as well. We implement the proposed method into a tool called ALPHAPROG. We analyze the framework with four different reward functions and our study reveal the effectiveness of ALPHAPROG for compiler testing. We also reported two important bugs for a compiler production that were confirmed and addressed by the project owner, which further demonstrates ALPHAPROG's applied value in practice.

poster session 7 @ red 6, poster session 8 @ red 6, poster session 7, poster session 8

Predictive VM (Virtual Machine) auto-scaling is a promising technique to optimize cloud applications’ operating costs and performance. Understanding the job arrival rate is crucial for accurately predicting future changes in cloud workloads and proactively provisioning and de-provisioning VMs for hosting the applications. However, developing a model that accurately predicts cloud workload changes is extremely challenging due to the dynamic nature of cloud workloads. Long- Short-Term-Memory (LSTM) models have been developed for cloud workload prediction. Unfortunately, the state-of-the-art LSTM model leverages recurrences to predict, which naturally adds complexity and increases the inference overhead as input sequences grow longer. To develop a cloud workload prediction model with high accuracy and low inference overhead, this work presents a novel time-series forecasting model called WGAN-gp Transformer, inspired by the Transformer network and improved Wasserstein-GANs. The proposed method adopts a Transformer network as a generator and a multi-layer perceptron as a critic. The extensive evaluations with real-world workload traces show WGAN- gp Transformer achieves 5× faster inference time with up to 5.1% higher prediction accuracy against the state-of-the-art. We also apply WGAN-gp Transformer to auto-scaling mechanisms on Google cloud platforms, and the WGAN-gp Transformer-based auto-scaling mechanism outperforms the LSTM-based mechanism by significantly reducing VM over-provisioning and under-provisioning rates.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7

Machine learning based image classification algorithms, such as deep neural network approaches, will be increasingly employed in critical settings such as quality control in industry, where transparency and comprehensibility of decisions are crucial. Therefore, we aim to extend the defect detection task towards an interactive human-in-the-loop approach that allows us to integrate rich background knowledge and the inference of complex relationships going beyond traditional purely data-driven approaches. We propose an approach for an interactive support system for classifications in an industrial quality control setting that combines the advantages of both (explainable) knowledge-driven and data-driven machine learning methods, in particular inductive logic programming and convolutional neural networks, with human expertise and control. The resulting system can assist domain experts with decisions, provide transparent explanations for results, and integrate feedback from users; thus reducing workload for humans while both respecting their expertise and without removing their agency or accountability.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7

We present a machine-learning-guided process that can efficiently extract factor tables from unstructured rate filing documents. Our approach combines multiple deep-learning-based models that work in tandem to create structured representations of tabular data present in unstructured documents such as pdf files. This process combines CNN's to detect tables, language-based models to extract table metadata and conventional computer vision techniques to improve the accuracy of tabular data on the machine-learning side. The extracted tabular data is validated through an intuitive user interface. This process, which we call Harvest, significantly reduces the time needed to extract tabular information from PDF files, enabling analysis of such data at a speed and scale that was previously unattainable.

poster session 7 @ red 6, poster session 8 @ red 6, poster session 7, poster session 8

Formal response organizations perform rapid damage assessments after natural and human-induced disasters to measure the extent of damage to infrastructures such as roads, bridges, and buildings. This time-critical task, when performed using traditional approaches such as experts surveying the disaster areas, poses serious challenges and delays response. This paper presents an AI-based system that leverages citizen science to collect damage images reported on social media and perform rapid damage assessment in real-time. Several image processing models in the system tackle non-trivial challenges posed by social media as a data source, such as high-volume of redundant and irrelevant content. The system determines the severity of damage using a state-of-the-art computer vision model. Together with a response organization in the US, we deployed the system to identify damage reports during a major real-world disaster. We observe that almost 42% of the images are unique, 28% relevant, and more importantly, only 10% of them contain either mild or severe damage. Experts from our partner organization provided feedback on the system's mistakes, which we used to perform additional experiments to retrain the models. Consequently, the retrained models based on expert feedback on the target domain data helped us achieve significant performance improvements.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7

Finding or tracking the location of an object accurately is a crucial problem in defense applications, robotics and computer vision. Radars fall into the spectrum of high-end defense sensors or systems upon which the security and surveillance of the entire world depends. There has been a lot of focus on the topic of Multi Sensor Tracking in recent years, with radars as the sensors. The Indian Air Force uses a Multi Sensor Tracking (MST) system to detect flights pan India, developed and supported by BEL(Bharat Electronics Limited), a defense agency we are working with. In this paper, we describe our Machine Learning approach, which is built on top of the existing system, the Air force uses. For purposes of this work, we trained our models on about 13 million anonymized real Multi Sensor tracking data points provided by radars performing tracking activity across the Indian air space. The approach has shown an increase in the accuracy of tracking by 5 percent from 91 to 96. The model and the corresponding code were transitioned to BEL, which has been tested in their simulation environment with a plan to take forward for ground testing. Our approach comprises of 3 steps: (a) We train a Neural Network model and a CatBoost model and ensemble them using a Logistic Regression model to predict one type of error, namely Splitting error, which can help to improve the accuracy of tracking. (b) We again train a Neural Network model and a CatBoost model and ensemble them using a different Logistic Regression model to predict the second type of error, namely Merging error, which can further improve the accuracy of tracking. (c) We use cosine similarity to find the nearest neighbour and correct the data points, predicted to have Splitting/Merging errors, by predicting the original global track of these data points.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7

Webscale services dealing with sensitive content are increasingly being deployed in public and hybrid cloud environments. At the same time, the impact of security breaches have also increased manifold averaging at USD 3.86M per data breach. To tackle such increasing risks, regulations and security frameworks are defined that an organization must comply with. Most of these frameworks are published in natural language text that run into hundreds of pages resulting into thousands of requirements and controls. When these frameworks undergo revisions, understanding the changes, and interpreting their impact consumes huge amount of time, effort and resources. In this paper, we propose a change management system that supports SMEs with AI-assisted automation of this extremely manual and time consuming activity. Specifically, we introduce the concept of live crosswalks – a framework that models complex relationships among security and compliance documents along with associated operations to manage the change. It uses natural language processing (NLP) and algorithmic techniques to transform the current document-driven, highly manual process into a data-driven interactive intelligent system. We present the overall design and demonstrate its efficacy over several hundreds of diversified controls through experimental evaluation.

poster session 7 @ red 6, poster session 8 @ red 6, poster session 7, poster session 8

Site Reliability Engineers (SREs) play a key role in identifying the cause of an issue and preforming remediation steps to resolve it. After an issue is reported, SREs come together in a virtual room (collaboration platform) to triage the issue. While doing so, they leave behind a wealth of information, in the form of conversations, which can be used later for triaging similar issues. However, usability of these conversations offer challenges due to them being and scarcity of conversation utterance label. This paper presents a novel approach for issue artefact extraction from noisy conversations with minimal labelled data. We propose a combination of unsupervised and supervised models with minimal human intervention that leverages domain knowledge to predict artefacts for a small amount of conversation data and use that for fine-tuning an already pre-trained language model for artefact prediction on a large amount of conversation data. Experimental results on our dataset show that the proposed ensemble of the unsupervised and supervised models is better than using either one of them individually. We also present a deployment case study of the proposed artefact prediction.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7

We approach the task of network congestion control in datacenters using Reinforcement Learning (RL). Successful congestion control algorithms can dramatically improve latency and overall network throughput. Until today, no such learning-based algorithms have shown practical potential in this domain. Evidently, the most popular recent deployments rely on rule-based heuristics that are tested on a predetermined set of benchmarks. Consequently, these heuristics do not generalize well to newly-seen scenarios. Contrarily, we devise an RL-based algorithm with the aim of generalizing to different configurations of real-world datacenter networks. We overcome challenges such as partial-observability, non-stationarity, and multi-objectiveness. We further propose a policy gradient algorithm that leverages the analytical structure of the reward function to approximate its derivative and improve stability. We show that these challenges prevent standard RL algorithms from operating within this domain. Our experiments, conducted on a realistic simulator that emulates communication networks' behavior, show that our method exhibits improved performance concurrently on the multiple considered metrics compared to the popular algorithms deployed today in real datacenters. Our algorithm is being productized to replace heuristics in some of the largest datacenters in the world.

poster session 9 @ red 6, poster session 7 @ red 6, poster session 9, poster session 7