We study the use of a time series encoder to learn representations that are useful on data set types with which it has not been trained on. The encoder is formed of a convolutional neural network whose temporal output is summarized by a convolutional attention mechanism. This way, we obtain a compact, fixed-length representation from longer, variable-length time series. We evaluate the performance of the proposed approach on a well-known time series classification benchmark, considering full adaptation, partial adaptation, and no adaptation of the encoder to the new data type. Results show that such strategies are competitive with the state-of-the-art, often outperforming conceptually-matching approaches. Besides accuracy scores, the facility of adaptation and the efficiency of pre-trained encoders make them an appealing option for the processing of scarcely- or non-labeled time series.

We report on our first experiments to train deep neural networks that automatically translate informalized $\LaTeX{}$-written Mizar texts into the formal Mizar language. Using Luong et al.'s neural machine translation model (NMT), we tested our aligned informal-formal corpora against various hyperparameters and evaluated their results. Our experiments show that NMT is able to generate correct Mizar statements on more than 60 percent of the inference data, indicating that formalization through artificial neural network is a promising approach for automated formalization of mathematics. We present several case studies to illustrate our results.

Can a robot grasp an unknown object without seeing it? In this paper, we present a tactile-sensing based approach to this challenging problem of grasping novel objects without prior knowledge of their location or physical properties. Our key idea is to combine touch based object localization with tactile based re-grasping. To train our learning models, we created a large-scale grasping dataset, including more than 30 RGB frames and over 2.8 million tactile samples from 7800 grasp interactions of 52 objects. To learn a representation of tactile signals, we propose an unsupervised auto-encoding scheme, which shows a significant improvement of 4-9% over prior methods on a variety of tactile perception tasks. Our system consists of two steps. First, our touch localization model sequentially 'touch-scans' the workspace and uses a particle filter to aggregate beliefs from multiple hits of the target. It outputs an estimate of the object's location, from which an initial grasp is established. Next, our re-grasping model learns to progressively improve grasps with tactile feedback based on the learned features. This network learns to estimate grasp stability and predict adjustment for the next grasp. Re-grasping thus is performed iteratively until our model identifies a stable grasp. Finally, we demonstrate extensive experimental results on grasping a large set of novel objects using tactile sensing alone. Furthermore, when applied on top of a vision-based policy, our re-grasping model significantly boosts the overall accuracy by 10.6%. We believe this is the first attempt at learning to grasp with only tactile sensing and without any prior object knowledge.

In recent years, AI researchers have developed and fine-tuned deep-learning networks -- layered programs that can come up with novel solutions to achieve their assigned goal. For example, a deep-learning network can be told which face to identify in a series of different photos, and through several rounds of training, can tune its algorithms until it spots the right face virtually every time.

Machine learning algorithms typically search for the optimal representation of data using some feedback signal (aka objective/loss function). However, most machine learning algorithms only have the ability to use one or two layers of data transformation to learn the output representation. As data sets continue to grow in the dimensions of the feature space, finding the optimal output representation with a shallow model is not always possible. Deep learning provides a multi-layer approach to learn data representations, typically performed with a multi-layer neural network. Like other machine learning algorithms, deep neural networks (DNN) perform learning by mapping features to targets through a process of simple data transformations and feedback signals; however, DNNs place an emphasis on learning successive layers of meaningful representations.

Deep Learning made easy with Deep Cognition This past month I had the luck to meet the founders of DeepCognition.ai. Deep Cognition breaks the significant barrier…becominghuman.ai Dataset for this problem has been collected by researcher at Case Western Reserve University in Cleveland, Ohio. Original dataset is available here (Edit: the original link is not working anymore, download from Kaggle). This dataset is preprocessed by nice people at Kaggle that was used as starting point in our work.

Machine learning, neural networks, deep learning, and artificial intelligence are all around us, and they're not going away. I will show you how to get a grasp on this ever-growing technology in this course. This course was funded by a wildly successful Kickstarter! With this course I will help you understand what machine learning is and compare it to Artificial Intelligence (AI). Together we will discover applications of machine learning and where we use machine learning daily.

Notch up another win for the robots: the latest program from Google's artificial intelligence group, DeepMind, has trounced experts at a maze game after it learned to find its way around like a human. Scientists noticed that when they trained the AI to move through a landscape, it spontaneously developed electrical activity akin to that seen in the specialised brain cells that underpin human navigational skills. So-called'grid cells' were only identified in animals in 2005 in work that earned researchers a Nobel prize. The latest breakthrough reveals the potential for human brain-like activity to emerge from scratch in AI systems. Beyond making smarter programs, it paves the way for computer engineers to build models that help neuroscientists better understand the human brain.

After 90 days and 288,000 working hours, the much-discussed fourth annual Data Science Bowl has ended. Run by Booz Allen Hamilton and Kaggle, the contest resulted in 68,000 algorithms, 3 winners, and one tantalizing opportunity for biomedical research. The goal of this year's Data Science Bowl was to build artificial intelligence (AI) systems that could automate what organizers called a "critical component of biomedical research." As such, 18,000 competitors spent months honing deep-learning models to scrutinize images of cells in search of nuclei, all without aid from humans. The ensuing algorithms are expected to salvage hundreds of thousands of hours each year, time that was previously burned by researchers who were forced to perform the task, according to the organizers.

In response to cyber-defenders' increasing use of AI technologies, malicious actors are discussing their potential application for criminal use. Research from Control Risks, the specialist global risk consultancy, shows that cyber-threat actors are actively exploring the development of innovative new techniques to use these technologies and tools to enhance their capabilities. For instance, in the post-infection phase, clusters of compromised devices, dubbed hivenets, could develop the ability to self-learn and could be used to automatically identify and target additional vulnerable systems. "More and more organizations are beginning to employ machine learning and artificial intelligence as part of their defenses against cyber-threats," said Nicolas Reys, associate director and head of the Control Risk cyber-threat intelligence team. "Cyber-threat actors are recognizing the need to advance their skills to keep up with this development. One application could be to use deep learning algorithms to improve the effectiveness of their attacks. This shows that AI and its subsets will play a larger role in facilitating cyber-attacks in the near future."