Deep convolutional neural networks for multi-scale time-series classification and application to disruption prediction in fusion devices R.M. Churchill Theory Department Princeton Plasma Physics Laboratory 100 Stellarator Road, Princeton, NJ 08540, USA rchurchi@pppl.gov and the DIII-D team General Atomics P .O. Box 85608, San Diego, California 92186, USA Abstract The multi-scale, mutli-physics nature of fusion plasmas makes predicting plasma events challenging. Recent advances in deep convolutional neural network architectures (CNN) utilizing dilated convolutions enable accurate predictions on sequences which have long-range, multi-scale characteristics, such as the time-series generated by diagnostic instruments observing fusion plasmas. Here we apply this neural network architecture to the popular problem of disruption prediction in fusion tokamaks, utilizing raw data from a single diagnostic, the Electron Cyclotron Emission imaging (ECEi) diagnostic from the DIII-D tokamak. ECEi measures a fundamental plasma quantity (electron temperature) with high temporal resolution over the entire plasma discharge, making it sensitive to a number of potential pre-disruptions markers with different temporal and spatial scales. Promising, initial disruption prediction results are obtained training a deep CNN with large receptive field ( 30k), achieving an F 1-score of 91% on individual time-slices using only the ECEi data. 1 Introduction Plasma phenomena contain a wide range of temporal and spatial scales, often exhibiting multi-scale characteristics (see Figure 1).

The recorded traffic was aggregated into 100x100 meters bins and made available as three-channel images. Within these images, the first channel depicts the traffic volume in each cell, the second one the average speed of vehicles, and the third one the majority of vehicles' directions (as one of four cardinal directions). The data spanned a whole year in 5-minute intervals, where certain days were left out from the training data, to be used for prediction and upload to the traffic4cast servers, which then would assess the quality of the prediction. The prediction itself consisted of "three images into the future" (spanning a 15-minute interval), based on the previous hour (12 images). Given the problem formalization, our efforts mostly focused on the application of well-known image processing algorithms, though we also explored various simple baselines, neural networks taking into account spatiotemporal context, as well as more complex network architectures that should be able to take advantage of the fact that the origin of the data stems from probes that move on a known graph. Ultimately, we did not manage to outperform the "simple" application of a widely-used image processing algorithm, which might be a hint that either a lot more research on networks targeted specifically at this problem or a different formulation of the problem altogether is required.

Learning representations for feature interactions to model user behaviors is critical for recommendation system and click-trough rate (CTR) predictions. Recent advances in this area are empowered by deep learning methods which could learn sophisticated feature interactions and achieve the state-of-the-art result in an end-to-end manner. These approaches require large number of training parameters integrated with the low-level representations, and thus are memory and computational inefficient. In this paper, we propose a new model named "LorentzFM" that can learn feature interactions embedded in a hyperbolic space in which the violation of triangle inequality for Lorentz distances is available. To this end, the learned representation is benefited by the peculiar geometric properties of hyperbolic triangles, and result in a significant reduction in the number of parameters (20\% to 80\%) because all the top deep learning layers are not required. With such a lightweight architecture, LorentzFM achieves comparable and even materially better results than the deep learning methods such as DeepFM, xDeepFM and Deep \& Cross in both recommendation and CTR prediction tasks.

Learning how to automatically solve optimization problems has the potential to provide the next big leap in optimization technology. The performance of automatically learned heuristics on routing problems has been steadily improving in recent years, but approaches based purely on machine learning are still outperformed by state-of-the-art optimization methods. To close this performance gap, we propose a novel large neighborhood search (LNS) framework for vehicle routing that integrates learned heuristics for generating new solutions. The learning mechanism is based on a deep neural network with an attention mechanism and has been especially designed to be integrated into an LNS search setting. We evaluate our approach on the capacitated vehicle routing problem (CVRP) and the split delivery vehicle routing problem (SDVRP). On CVRP instances with up to 297 customers our approach significantly outperforms an LNS that uses only handcrafted heuristics and a well-known heuristic from the literature. Furthermore, we show for the CVRP and the SDVRP that our approach surpasses the performance of existing machine learning approaches and comes close to the performance of state-of-the-art optimization approaches.

What Do Y ou Mean'Why?': Resolving Sluices in Conversations Victor Petr en Bach Hansen, 1 2 Anders Søgaard 1 3 1 Department of Computer Science, University of Copenhagen, Denmark 2 Topdanmark A/S, Denmark 3 Google Research, Berlin victor.petren@di.ku.dk, soegaard@di.ku.dk Abstract In conversation, we often ask one-word questions such as'Why?' or'Who?'. Such questions are typically easy for humans to answer, but can be hard for computers, because their resolution requires retrieving both the right semantic frames and the right arguments from context. This paper introduces the novel ellipsis resolution task of resolving such one-word questions, referred to as sluices in linguistics. We present a crowd-sourced dataset containing annotations of sluices from over 4,000 dialogues collected from conversational QA datasets, as well as a series of strong baseline architectures. 1 Introduction Stand-alone wh-word questions, such as When? in Figure 1, are easy for us to understand, but in order to interpret them we need to retrieve implicit information from context. Learning to do so is an instance of sluicing, an ellipsis phenomenon, defined by Ross (1969) as'the effect of deleting everything but the preposed constituent of an embedded question, under the condition that the remainder of the question is identical to some other part of the sentence, or a preceding sentence.' In the context of conversations, one-word wh-word questions are particularly frequent (Anand and Hardt 2016; Rønning, Hardt, and Søgaard 2018), and because they are often hard to resolve, they seem to be a frequent source of error in conversational question answering (Choi et al. 2018; Reddy, Chen, and Manning 2018) and dialogue understanding (Vlachos and Clark 2014). We refer to this type of sluicing as conversational sluicing . Unlike previous work where sluice resolution is treated as predicting the span of the antecedent (Anand and Hardt 2016; Rønning, Hardt, and Søgaard 2018), we frame conversational sluice resolution as a Natural Language Generation (NLG) task, in which we seek to automatically generate the full question, given a question-answer context and a one-word question. Q 1: Where was the bombing?

One of the most ambitious use cases of computer-assisted learning is to build a lifelong learning recommendation system. Unlike short-term courses, lifelong learning presents unique challenges, requiring sophisticated recommendation models that account for a wide range of factors such as background knowledge of learners or novelty of the material while effectively maintaining knowledge states of masses of learners for significantly longer periods of time (ideally, a lifetime). This work presents the foundations towards building a dynamic, scalable and transparent recommendation system for education, modelling learner's knowledge from implicit data in the form of engagement with open educational resources. We i) use a text ontology based on Wikipedia to automatically extract knowledge components of educational resources and, ii) propose a set of online Bayesian strategies inspired by the well-known areas of item response theory and knowledge tracing. Our proposal, TrueLearn, focuses on recommendations for which the learner has enough background knowledge (so they are able to understand and learn from the material), and the material has enough novelty that would help the learner improve their knowledge about the subject and keep them engaged. We further construct a large open educational video lectures dataset and test the performance of the proposed algorithms, which show clear promise towards building an effective educational recommendation system. Introduction One-on-one tutoring has shown learning gains of the order of two standard deviations (Corbett 2001). Machine learning now promises to provide such benefits of high quality personalised teaching to anyone in the world in a cost effective manner (Piech et al. 2015). Meanwhile, Open Educational Resources (OERs), defined as teaching, learning and research material available in the public domain or published under an open license (UNESCO 2019), are growing at a very fast pace.

Previous works related to automatic personality recognition focus on using traditional classification models with linguistic features. However, attentive neural networks with contextual embeddings, which have achieved huge success in text classification, are rarely explored for this task. In this project, we have two major contributions. First, we create the first dialogue-based personality dataset, FriendsPersona, by annotating 5 personality traits of speakers from Friends TV Show through crowdsourcing. Second, we present a novel approach to automatic personality recognition using pre-trained contextual embeddings (BERT and RoBERTa) and attentive neural networks. Our models largely improve the state-of-art results on the monologue Essays dataset by 2.49%, and establish a solid benchmark on our FriendsPersona. By comparing results in two datasets, we demonstrate the challenges of modeling personality in multi-party dialogue.

CAIML #9 took place on November 14 at factor-a – part of Dept, demonstrating how AI can be used for social good and to address societal challenges. "Aid organizations and governments are applying great effort in resolving the negative impacts of food insecurity induced crisis like famines or mass migration. One of the most limiting resources these actors face is the lack of preparation time for consistent and sustainable planning for emergency relief like setting refugee camps or securing supply with food and energy. Hence, increasing the lead time for preparation is an essential step and will result in saving many lives. The aim of this research is to increase the lead time by developing a ML based mathematical prediction model that is able to compute the probability for food insecure areas by learning from historical data. For performing such computations, our prediction model is developed and trained on historic open access data for the Horn of Africa (2009-2018). We used precipitation and vegetation data derived by remote sensing, as well as socio-economic, medical, armed conflict and disaster data. To overcome spatial inconsistencies in the input data and to meet the requirements of spatially homogenous input for neural networks, all data has been converted to geo-referenced raster maps. Disaster and armed conflict data has been fitted to districts while local food market prices have been interpolated. The IPC has been used as the food security label. In order to find a prediction model, deep learning methods have been used. Several analyses were applied on the collected data such as multicollinearity checks and principal component analyses. Preliminary cross-validated results have encouraged us to further investigate the detection of food insecure areas using open access data."

We discuss some wonders in the field of image processing with machine learning advancements. Image processing can be defined as the technical analysis of an image by using complex algorithms. Here, image is used as the input, where the useful information returns as the output. According to a report, the image processing industry will reach USD 38.9 billion by 2021. Meanwhile, the Artificial Intelligence industry is also turning to a considerable growth curve.

We started collaborating with the Play store to help develop and improve systems that determine the relevance of an app with respect to the user. In this post, we'll explore some of the cutting-edge machine learning techniques we developed to achieve this. Today, Google Play's recommendation system contains three main models: a candidate generator, a reranker, and a model to optimise for multiple objectives. The candidate generator is a deep retrieval model that can analyse more than a million apps and retrieve the most suitable ones. For each app, a reranker, i.e. a user preference model, predicts the user's preferences along multiple dimensions.