Detecting epileptic seizure through analysis of the electroencephalography (EEG) signal becomes a standard method for the diagnosis of epilepsy. In a manual way, monitoring of long term EEG is tedious and error prone. Therefore, a reliable automatic seizure detection method is desirable. A critical challenge to automatic seizure detection is that seizure morphologies exhibit considerable variabilities. In order to capture essential seizure patterns, this paper leverages an attention mechanism and a bidirectional long short-term memory (BiLSTM) model to exploit both spatially and temporally discriminating features and account for seizure variabilities. The attention mechanism is to capture spatial features more effectively according to the contributions of brain areas to seizures. The BiLSTM model is to extract more discriminating temporal features in the forward and the backward directions. By accounting for both spatial and temporal variations of seizures, the proposed method is more robust across subjects. The testing results over the noisy real data of CHB-MIT show that the proposed method outperforms the current state-of-the-art methods. In both mixing-patients and cross-patient experiments, the average sensitivity and specificity are both higher while their corresponding standard deviations are lower than the methods in comparison.

What has happened in machine learning lately, and what does it mean for the future of medical image analysis? Machine learning has witnessed a tremendous amount of attention over the last few years. The current boom started around 2009 when so-called deep artificial neural networks began outperforming other established models on a number of important benchmarks. Deep neural networks are now the state-of-the-art machine learning models across a variety of areas, from image analysis to natural language processing, and widely deployed in academia and industry. These developments have a huge potential for medical imaging technology, medical data analysis, medical diagnostics and healthcare in general, slowly being realized. We provide a short overview of recent advances and some associated challenges in machine learning applied to medical image processing and image analysis. As this has become a very broad and fast expanding field we will not survey the entire landscape of applications, but put particular focus on deep learning in MRI. Our aim is threefold: (i) give a brief introduction to deep learning with pointers to core references; (ii) indicate how deep learning has been applied to the entire MRI processing chain, from acquisition to image retrieval, from segmentation to disease prediction; (iii) provide a starting point for people interested in experimenting and perhaps contributing to the field of machine learning for medical imaging by pointing out good educational resources, state-of-the-art open-source code, and interesting sources of data and problems related medical imaging.

Human activity recognition based on video streams has received numerous attentions in recent years. Due to lack of depth information, RGB video based activity recognition performs poorly compared to RGB-D video based solutions. On the other hand, acquiring depth information, inertia etc. is costly and requires special equipment, whereas RGB video streams are available in ordinary cameras. Hence, our goal is to investigate whether similar or even higher accuracy can be achieved with RGB-only modality. In this regard, we propose a novel framework that couples skeleton data extracted from RGB video and deep Bidirectional Long Short Term Memory (BLSTM) model for activity recognition. A big challenge of training such a deep network is the limited training data, and exploring RGB-only stream significantly exaggerates the difficulty. We therefore propose a set of algorithmic techniques to train this model effectively, e.g., data augmentation, dynamic frame dropout and gradient injection. The experiments demonstrate that our RGB-only solution surpasses the state-of-the-art approaches that all exploit RGB-D video streams by a notable margin. This makes our solution widely deployable with ordinary cameras.

Recently, deep neural networks (DNNs) have been widely applied in mobile intelligent applications. The inference for the DNNs is usually performed in the cloud. However, it leads to a large overhead of transmitting data via wireless network. In this paper, we demonstrate the advantages of the cloud-edge collaborative inference with quantization. By analyzing the characteristics of layers in DNNs, an auto-tuning neural network quantization framework for collaborative inference is proposed. We study the effectiveness of mixed-precision collaborative inference of state-of-the-art DNNs by using ImageNet dataset. The experimental results show that our framework can generate reasonable network partitions and reduce the storage on mobile devices with trivial loss of accuracy.

One of the core challenges in Visual Dialogue problems is asking the question that will provide the most useful information towards achieving the required objective. Encouraging an agent to ask the right questions is difficult because we don't know a-priori what information the agent will need to achieve its task, and we don't have an explicit model of what it knows already. We propose a solution to this problem based on a Bayesian model of the uncertainty in the implicit model maintained by the visual dialogue agent, and in the function used to select an appropriate output. By selecting the question that minimises the predicted regret with respect to this implicit model the agent actively reduces ambiguity. The Bayesian model of uncertainty also enables a principled method for identifying when enough information has been acquired, and an action should be selected. We evaluate our approach on two goal-oriented dialogue datasets, one for visual-based collaboration task and the other for a negotiation-based task. Our uncertainty-aware information-seeking model outperforms its counterparts in these two challenging problems.

Video Friday is your weekly selection of awesome robotics videos, collected by your Automaton bloggers. We'll also be posting a weekly calendar of upcoming robotics events for the next few months; here's what we have so far (send us your events!): Let us know if you have suggestions for next week, and enjoy today's videos. The bar has now been set for robot holiday videos, thanks to FZI. Still waiting for a robot with a cookie to show up at my door.

An increasing number of modern antivirus solutions rely on machine learning (ML) techniques to protect users from malware. While ML-based approaches, like FireEye Endpoint Security's MalwareGuard capability, have done a great job at detecting new threats, they also come with substantial development costs. Creating and curating a large set of useful features takes significant amounts of time and expertise from malware analysts and data scientists (note that in this context a feature refers to a property or characteristic of the executable that can be used to distinguish between goodware and malware). In recent years, however, deep learning approaches have shown impressive results in automatically learning feature representations for complex problem domains, like images, speech, and text. Can we take advantage of these advances in deep learning to automatically learn how to detect malware without costly feature engineering? As it turns out, deep learning architectures, and in particular convolutional neural networks (CNNs), can do a good job of detecting malware simply by looking at the raw bytes of Windows Portable Executable (PE) files.

When I published the article "Understanding Blockchain" many of you wrote me to ask me if I could make one dedicated to Artificial Intelligence. The truth is that I hadn't had time to get on with it and before sharing anything, I wanted to finish some courses in order to add value to the recommendations. The problem with Artificial Intelligence is that it's much more fragmented, both technologically and in use cases, than Blockchain, making it a real challenge to condense all the information and share it meaningfully. Likewise, I have tried to make an effort in the summary of key concepts and in the compilation of interesting sources and resources, I hope it helps you as well as it did to me! Let's start with a little history. The timeline you see is taken from this article and it shows the most important milestones of Artificial Intelligence.

Years ago, when I was taking my first steps in computer programming, coding was for geeks and computer programs had limited use. Development tools were very crude, writing code was hard (remember Assembly, C, Pascal?), compiling and linking was a nightmare (MAKE files anyone?), and debugging was even worse. Long story short, programming was not for the faint of heart. You needed nerves of steel and had to patiently fail over and over before you got the hang of writing good code. But as software gradually rose in prominence, the entry barrier to programming lowered.

State-of-the-art image recognition systems use sophisticated Convolutional Neural Networks (CNNs) that are designed and trained to identify numerous object classes. Such networks are fairly resource intensive to compute, prohibiting their deployment on resource-constrained embedded platforms. On one hand, the ability to classify an exhaustive list of categories is excessive for the demands of most IoT applications. On the other hand, designing a new custom-designed CNN for each new IoT application is impractical, due to the inherent difficulty in developing competitive models and time-to-market pressure. To address this problem, we investigate the question of: "Can one utilize an existing optimized CNN model to automatically build a competitive CNN for an IoT application whose objects of interest are a fraction of categories that the original CNN was designed to classify, such that the resource requirement is proportionally scaled down?" We use the term resource scalability to refer to this concept, and develop a methodology for automated synthesis of resource scalable CNNs from an existing optimized baseline CNN. The synthesized CNN has sufficient learning capacity for handling the given IoT application requirements, and yields competitive accuracy. The proposed approach is fast, and unlike the presently common practice of CNN design, does not require iterative rounds of training trial and error.