This paper presents a publicly available corpus of French encyclopedic history texts annotated according to the Berkeley FrameNet formalism. The main difference in our approach compared to previous works on semantic parsing with FrameNet is that we are not interested here in full text parsing but rather on partial parsing. The goal is to select from the FrameNet resources the minimal set of frames that are going to be useful for the applicative framework targeted, in our case Information Extraction from encyclopedic documents. Such an approach leverages the manual annotation of larger corpora than those obtained through full text parsing and therefore opens the door to alternative methods for Frame parsing than those used so far on the FrameNet 1.5 benchmark corpus. The approaches compared in this study rely on an integrated sequence labeling model which jointly optimizes frame identification and semantic role segmentation and identification.

The power of randomized algorithms in numerical methods have led to fast solutions which use the Singular Value Decomposition (SVD) as a core routine. However, given the large data size of modern and the modest runtime of SVD, most practical algorithms would require some form of approximation, such as sketching, when running SVD. While these approximation methods satisfy many theoretical guarantees, we provide the first algorithmic implementations for sketch-and-solve SVD problems on real-world, large-scale datasets. We provide a comprehensive empirical evaluation of these algorithms and provide guidelines on how to ensure accurate deployment to real-world data. As an application of sketched SVD, we present Sketched Leverage Score Ordering, a technique for determining the ordering of data in the training of neural networks. Our technique is based on the distributed computation of leverage scores using random projections. These computed leverage scores provide a flexible and efficient method to determine the optimal ordering of training data without manual intervention or annotations. We present empirical results on an extensive set of experiments across image classification, language sentiment analysis, and multi-modal sentiment analysis. Our method is faster compared to standard randomized projection algorithms and shows improvements in convergence and results.

In an interdisciplinary project funded by a Canadian Institute for Advanced Research (CIFAR) Catalyst grant, researchers at the University of Guelph and the University of Toronto, Mississauga combined expertise in fruit fly biology with machine learning to build a biologically-based algorithm that churns through low-resolution videos of fruit flies in order to test whether it is physically possible for a system with such constraints to accomplish such a difficult task. Fruit flies have small compound eyes that take in a limited amount of visual information, an estimated 29 units squared. The traditional view has been that once the image is processed by a fruit fly, it is only able to distinguish very broad features. But a recent discovery that fruit flies can boost their effective resolution with subtle biological tricks has led researchers to believe that vision could contribute significantly to the social lives of flies. This, combined with the discovery that the structure of their visual system looks a lot like a Deep Convolutional Network (DCN), led the team to ask: "can we model a fly brain that can identify individuals?"

The rate of progress in the field of artificial intelligence is one of the most hotly contested aspects of the ongoing boom in teaching computers and robots how to see the world, make sense of it, and eventually perform complex tasks both in the physical realm and the virtual one. And just how fast the industry is moving, and to what end, is typically measured not just by actual product advancements and research milestones, but also by the prognostications and voiced concerns of AI leaders, futurists, academics, economists, and policymakers. AI is going to change the world -- but how and when are still open questions. Today, findings from a group of experts were published in an ongoing effort to help answer those questions. The experts include members of Harvard, MIT, Stanford, the nonprofit OpenAI, and the Partnership on AI industry consortium, among others, and they were put together as part of the second annual AI Index.

Abstract--Hybrid LSTM-fully convolutional networks (LSTM-FCN) for time series classification have produced state-of-the-art classification results on univariate time series. We show that replacing the LSTM with a gated recurrent unit (GRU) to create a GRU-fully convolutional network hybrid model (GRU-FCN) can offer even better performance on many time series datasets. The proposed GRU-FCN model outperforms state-of-the-art classification performance in many univariate and multivariate time series datasets. In addition, since the GRU uses a simpler architecture than the LSTM, it has fewer training parameters, less training time, and a simpler hardware implementation, compared to the LSTM-based models. A time series (TS) is a sequence of data points obtained at successive equally-spaced time points, ordinarily in a uniform interval time domain [1].

Smart mobility management would be an important prerequisite for future fog computing systems. In this research, we propose a learning-based handover optimization for the Internet of Vehicles that would assist the smooth transition of device connections and offloaded tasks between fog nodes. To accomplish this, we make use of machine learning algorithms to learn from vehicle interactions with fog nodes. Our approach uses a three-layer feed-forward neural network to predict the correct fog node at a given location and time with 99.2 % accuracy on a test set. We also implement a dual stacked recurrent neural network (RNN) with long short-term memory (LSTM) cells capable of learning the latency, or cost, associated with these service requests. We create a simulation in JAMScript using a dataset of real-world vehicle movements to create a dataset to train these networks. We further propose the use of this predictive system in a smarter request routing mechanism to minimize the service interruption during handovers between fog nodes and to anticipate areas of low coverage through a series of experiments and test the models' performance on a test set.

We consider the problem of sufficient dimensionality reduction (SDR), where the high-dimensional observation is transformed to a low-dimensional sub-space in which the information of the observations regarding the label variable is preserved. We propose DVSDR, a deep variational approach for sufficient dimensionality reduction. The deep structure in our model has a bottleneck that represent the low-dimensional embedding of the data. We explain the SDR problem using graphical models and use the framework of variational autoencoders to maximize the lower bound of the log-likelihood of the joint distribution of the observation and label. We show that such a maximization problem can be interpreted as solving the SDR problem. DVSDR can be easily adopted to semi-supervised learning setting. In our experiment we show that DVSDR performs competitively on classification tasks while being able to generate novel data samples.

Automatic summarization of natural language is a current topic in computer science research and industry, studied for decades because of its usefulness across multiple domains. For example, summarization is necessary to create reviews such as this one. Research and applications have achieved some success in extractive summarization (where key sentences are curated), however, abstractive summarization (synthesis and re-stating) is a hard problem and generally unsolved in computer science. This literature review contrasts historical progress up through current state of the art, comparing dimensions such as: extractive vs. abstractive, supervised vs. unsupervised, NLP (Natural Language Processing) vs Knowledge-based, deep learning vs algorithms, structured vs. unstructured sources, and measurement metrics such as Rouge and BLEU. Multiple dimensions are contrasted since current research uses combinations of approaches as seen in the review matrix. Throughout this summary, synthesis and critique is provided. This review concludes with insights for improved abstractive summarization measurement, with surprising implications for detecting understanding and comprehension in general.

Predicting the behavior of surrounding vehicles is a critical problem in automated driving. We present a novel game theoretic behavior prediction model that achieves state of the art prediction accuracy by explicitly reasoning about possible future interaction between agents. We evaluate our approach on the NGSIM vehicle trajectory data set and demonstrate lower root mean square error than state-of-the-art methods.

Understanding the loss surface of neural networks is essential for the design of models with predictable performance and their success in applications. Experimental results suggest that sufficiently deep and wide neural networks are not negatively impacted by suboptimal local minima. Despite recent progress, the reason for this outcome is not fully understood. Could deep networks have very few, if at all, suboptimal local optima? or could all of them be equally good? We provide a construction to show that suboptimal local minima (i.e. non-global ones), even though degenerate, exist for fully connected neural networks with sigmoid activation functions. The local minima obtained by our proposed construction belong to a connected set of local solutions that can be escaped from via a non-increasing path on the loss curve. For extremely wide neural networks with two hidden layers, we prove that every suboptimal local minimum belongs to such a connected set. This provides a partial explanation for the successful application of deep neural networks. In addition, we also characterize under what conditions the same construction leads to saddle points instead of local minima for deep neural networks.