One way to measure progress in artificial intelligence is to chart victories by algorithms over champions of increasingly challenging games--checkers, chess, and, in 2016, Go. On Wednesday, five bots sought to extend AI's mastery to e-sports, in the fantasy battle game Dota 2. They failed, as a team of pro gamers from Brazil called paiN defended humanity's honor--for now. A crowd of thousands in Vancouver's hockey arena watched the bots battle paiN over 52 tense minutes packed with spells and firebolts. The human-machine contest was a side event to The International, a Dota 2 tournament that boasts the biggest purse in e-sports, at $25 million. The five bots that lost Wednesday were created by OpenAI, a research institute cofounded by Tesla CEO Elon Musk to work towards human-level artificial intelligence, and make the technology safe.

The International In the past hour, OpenAI's artificially intelligent bots lost their first match against professional players at smash-hit computer game Dota 2 at The International – the video game's annual championship tournament. It's the first bout in a best-of-three competition between human professional players versus OpenAI's code, the other two rounds will take place over the next two days, each day a different human team. Thousands of hardcore Dota fans lit up by glowing bracelets sat down at the Rogers Arena in Vancouver, Canada, to watch pros battle against a machine running OpenAI's software in this first round. The humans – dubbed Team paiN – were five players from Brazil, while OpenAI Five is made up five long-short-term memory neural-network-based agents. Dota 2 is a popular battle strategy game played online.

TensorFlow.js is a new version of the popular open-source library which brings deep learning to JavaScript. Developers can now define, train, and run machine learning models using the high-level library API. Pre-trained models mean developers can now easily perform complex tasks like visual recognition, generating music or detecting human poses with just a few lines of JavaScript. Having started as a front-end library for web browsers, recent updates added experimental support for Node.js. This allows TensorFlow.js to be used in backend JavaScript applications without having to use Python.

The Jacobian-based Saliency Map Attack is a family of adversarial attack methods for fooling classification models, such as deep neural networks for image classification tasks. By saturating a few pixels in a given image to their maximum or minimum values, JSMA can cause the model to misclassify the resulting adversarial image as a specified erroneous target class. We propose two variants of JSMA, one which removes the requirement to specify a target class, and another that additionally does not need to specify whether to only increase or decrease pixel intensities. Our experiments highlight the competitive speeds and qualities of these variants when applied to datasets of hand-written digits and natural scenes.

Reinforcement learning approaches have long appealed to the data management community due to their ability to learn to control dynamic behavior from raw system performance. Recent successes in combining deep neural networks with reinforcement learning have sparked significant new interest in this domain. However, practical solutions remain elusive due to large training data requirements, algorithmic instability, and lack of standard tools. In this work, we introduce LIFT, an end-to-end software stack for applying deep reinforcement learning to data management tasks. While prior work has frequently explored applications in simulations, LIFT centers on utilizing human expertise to learn from demonstrations, thus lowering online training times. We further introduce TensorForce, a TensorFlow library for applied deep reinforcement learning exposing a unified declarative interface to common RL algorithms, thus providing a backend to LIFT. We demonstrate the utility of LIFT in two case studies in database compound indexing and resource management in stream processing. Results show LIFT controllers initialized from demonstrations can outperform human baselines and heuristics across latency metrics and space usage by up to 70%.

Important advances have been made using convolutional neural network (CNN) approaches to solve complicated problems in areas that rely on grid structured data such as image processing and object classification. Recently, research on graph convolutional neural networks (GCNN) has increased dramatically as researchers try to replicate the success of CNN for graph structured data. Unfortunately, traditional CNN methods are not readily transferable to GCNN, given the irregularity and geometric complexity of graphs. The emerging field of GCNN is further complicated by research papers that differ greatly in their scope, detail, and level of academic sophistication needed by the reader. The present paper provides a review of some basic properties of GCNN. As a guide to the interested reader, recent examples of GCNN research are then grouped according to techniques that attempt to uncover the underlying topology of the graph model and those that seek to generalize traditional CNN methods on graph data to improve prediction of class membership. Discrete Signal Processing on Graphs (DSPg) is used as a theoretical framework to better understand some of the performance gains and limitations of these recent GCNN approaches. A brief discussion of Topology Adaptive Graph Convolutional Networks (TAGCN) is presented as an approach motivated by DSPg and future research directions using this approach are briefly discussed.

Abstract--Deep learning (DL), despite its enormous success in many computer vision and language processing applications, is exceedingly vulnerable to adversarial attacks. We consider the use of DL for radio signal (modulation) classification tasks, and present practical methods for the crafting of white-box and universal black-box adversarial attacks in that application. We show that these attacks can considerably reduce the classification performance, with extremely small perturbations of the input. In particular, these attacks are significantly more powerful than classical jamming attacks, which raises significant security and robustness concerns in the use of DLbased algorithms for the wireless physical layer. Deep learning (DL), implemented through deep neural networks (DNNs), represents a machine-learning paradigm that has been extremely successful in the last decade, especially in computer vision and natural language processing applications [1].

This paper studies the problems of vehicle make & model classification. Some of the main challenges are reaching high classification accuracy and reducing the annotation time of the images. To address these problems, we have created a fine-grained database using online vehicle marketplaces of Turkey. A pipeline is proposed to combine an SSD (Single Shot Multibox Detector) model with a CNN (Convolutional Neural Network) model to train on the database. In the pipeline, we first detect the vehicles by following an algorithm which reduces the time for annotation. Then, we feed them into the CNN model. It is reached approximately 4% better classification accuracy result than using a conventional CNN model. Next, we propose to use the detected vehicles as ground truth bounding box (GTBB) of the images and feed them into an SSD model in another pipeline. At this stage, it is reached reasonable classification accuracy result without using perfectly shaped GTBB. Lastly, an application is implemented in a use case by using our proposed pipelines. It detects the unauthorized vehicles by comparing their license plate numbers and make & models. It is assumed that license plates are readable.

Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder. Finding the biomarkers associated with ASD is extremely helpful to understand the underlying roots of the disorder and can lead to earlier diagnosis and more targeted treatment. Although Deep Neural Networks (DNNs) have been applied in functional magnetic resonance imaging (fMRI) to identify ASD, understanding the data-driven computational decision making procedure has not been previously explored. Therefore, in this work, we address the problem of interpreting reliable biomarkers associated with identifying ASD; specifically, we propose a 2-stage method that classifies ASD and control subjects using fMRI images and interprets the saliency features activated by the classifier. First, we trained an accurate DNN classifier. Then, for detecting the biomarkers, different from the DNN visualization works in computer vision, we take advantage of the anatomical structure of brain fMRI and develop a frequency-normalized sampling method to corrupt images. Furthermore, in the ASD vs. control subjects classification scenario, we provide a new approach to detect and characterize important brain features into three categories. The biomarkers we found by the proposed method are robust and consistent with previous findings in the literature. We also validate the detected biomarkers by neurological function decoding and comparing with the DNN activation maps.

In this work, we present a method to predict an entire `action tube' (a set of temporally linked bounding boxes) in a trimmed video just by observing a smaller subset of it. Predicting where an action is going to take place in the near future is essential to many computer vision based applications such as autonomous driving or surgical robotics. Importantly, it has to be done in real-time and in an online fashion. We propose a Tube Prediction network (TPnet) which jointly predicts the past, present and future bounding boxes along with their action classification scores. At test time TPnet is used in a (temporal) sliding window setting, and its predictions are put into a tube estimation framework to construct/predict the video long action tubes not only for the observed part of the video but also for the unobserved part. Additionally, the proposed action tube predictor helps in completing action tubes for unobserved segments of the video. We quantitatively demonstrate the latter ability, and the fact that TPnet improves state-of-the-art detection performance, on one of the standard action detection benchmarks - J-HMDB-21 dataset.