Translating natural language to SQL queries for table-based question answering is a challenging problem and has received significant attention from the research community. In this work, we extend a pointer-generator and investigate the order-matters problem in semantic parsing for SQL. Even though our model is a straightforward extension of a general-purpose pointer-generator, it outperforms early works for WikiSQL and remains competitive to concurrently introduced, more complex models. Moreover, we provide a deeper investigation of the potential order-matters problem that could arise due to having multiple correct decoding paths, and investigate the use of REINFORCE as well as a dynamic oracle in this context.

Anomaly detection in supercomputers is a very difficult problem due to the big scale of the systems and the high number of components. The current state of the art for automated anomaly detection employs Machine Learning methods or statistical regression models in a supervised fashion, meaning that the detection tool is trained to distinguish among a fixed set of behaviour classes (healthy and unhealthy states). We propose a novel approach for anomaly detection in High Performance Computing systems based on a Machine (Deep) Learning technique, namely a type of neural network called autoencoder. The key idea is to train a set of autoencoders to learn the normal (healthy) behaviour of the supercomputer nodes and, after training, use them to identify abnormal conditions. This is different from previous approaches which where based on learning the abnormal condition, for which there are much smaller datasets (since it is very hard to identify them to begin with). We test our approach on a real supercomputer equipped with a fine-grained, scalable monitoring infrastructure that can provide large amount of data to characterize the system behaviour. The results are extremely promising: after the training phase to learn the normal system behaviour, our method is capable of detecting anomalies that have never been seen before with a very good accuracy (values ranging between 88% and 96%).

This paper tackles a new problem setting: reinforcement learning with pixel-wise rewards (pixelRL) for image processing. After the introduction of the deep Q-network, deep RL has been achieving great success. However, the applications of deep RL for image processing are still limited. Therefore, we extend deep RL to pixelRL for various image processing applications. In pixelRL, each pixel has an agent, and the agent changes the pixel value by taking an action. We also propose an effective learning method for pixelRL that significantly improves the performance by considering not only the future states of the own pixel but also those of the neighbor pixels. The proposed method can be applied to some image processing tasks that require pixel-wise manipulations, where deep RL has never been applied. We apply the proposed method to three image processing tasks: image denoising, image restoration, and local color enhancement. Our experimental results demonstrate that the proposed method achieves comparable or better performance, compared with the state-of-the-art methods based on supervised learning.

Counterfactual Regret Minimization (CFR) is the leading algorithm for solving large imperfect-information games. It iteratively traverses the game tree in order to converge to a Nash equilibrium. In order to deal with extremely large games, CFR typically uses domain-specific heuristics to simplify the target game in a process known as abstraction. This simplified game is solved with tabular CFR, and its solution is mapped back to the full game. This paper introduces Deep Counterfactual Regret Minimization (Deep CFR), a form of CFR that obviates the need for abstraction by instead using deep neural networks to approximate the behavior of CFR in the full game. We show that Deep CFR is principled and achieves strong performance in large poker games. This is the first non-tabular variant of CFR to be successful in large games.

We present a novel algorithm to train a deep Q-learning agent using natural-gradient techniques. We compare the original deep Q-network (DQN) algorithm to its natural-gradient counterpart, which we refer to as NGDQN, on a collection of classic control domains. Without employing target networks, NGDQN significantly outperforms DQN without target networks, and performs no worse than DQN with target networks, suggesting that NGDQN stabilizes training and can help reduce the need for additional hyperparameter tuning. We also find that NGDQN is less sensitive to hyperparameter optimization relative to DQN. Together these results suggest that natural-gradient techniques can improve value-function optimization in deep reinforcement learning.

In 2014, Szegedy et al. published an ICLR paper with a surprising discovery: modern deep neural networks trained for image classification exhibit the following vulnerability: by making only slight alterations to an input image, it's possible to drastically fool a model that would otherwise classify the image correctly (say, as a dog), into outputting a completely wrong label (say, as a banana). Moreover, this attack is possible even with perturbations that are so tiny that a human couldn't distinguish the altered image from the original. These doctored images are called adversarial examples and the study of how to make neural networks robust to these attacks is an increasingly active area of machine learning research. This adversarial noise can be made robust. Labsix, a group of MIT students, showed that one can create adversarial examples in the physical world by 3d-printing a physical object, now affectionately called the adversarial turtle, that is classified as a rifle by state-of-the-art deep nets.

IMAGE: InSilico study reveals how E64 approaches, binds to, and inhibits falcipain-2 of Plasmodium falciparum that causes malaria in humans. Monday, November 12, 2018, Taipei, Taiwan, Republic of China - Insilico Taiwan, a Taipei-based subsidiary of Insilico Medicine, developing the end-to-end drug discovery pipeline utilizing the next generation artificial intelligence, announces the publication of a new research paper titled "In Silico Study Reveals How E64 Approaches, Binds to, and Inhibits Falcipain-2 of Plasmodium falciparum that Causes Malaria in Humans" in Scientific Reports - a scientific journal published by the Nature Publishing Group. Malaria is one of the world's oldest infectious diseases that still causes a lot of health problems in many tropical countries. Plasmodium falciparum, the most dangerous human malaria parasite, is believed to cause hundreds of millions of illnesses and about half a million deaths a year. Inhibitors of FP2 block haemoglobin destruction and parasite development, suggesting that FP2 inhibition is a promising target for antimalarial therapy.

There are several types of models that can be used for time-series forecasting. In this specific example, I used a Long short-term memory network, or in short LSTM Network, which is a special kind of neural network that make predictions according to the data of previous times. It is popular for language recognition, time series analysis and much more. However, in my experience, simpler types of models actually provide just as accurate predictions in many cases. Due to their sequential nature, TDNN's are implemented as a feedforward neural network instead of a recurrent neural network.

Artificial intelligence could spot early signs of Alzheimer's disease more than six years before a patient would normally be diagnosed, research suggests. Scientists conducting a small pilot study trained a self-learning computer programme to recognise features in brain scans which are too subtle for humans to see. The system was able to detect the beginnings of Alzheimer's in 40 patients an average of more than six years before they were formally diagnosed. The American researchers trained the "deep learning algorithm" using more than 2,100 PET (positron emission tomography) scans from 1,002 patients. PET scans measure metabolic activity in the brain by tracking the uptake of a radioactive glucose compound injected into the blood.

This article is a follow up to my previous one talking about the rise of Artificial Intelligence in an Enterprise. In this article, I will talk about how enterprises in Transportation, Retail and Healthcare are transforming themselves using AI. The use cases vary from transforming back-office applications to bringing compassion back into healthcare to detecting fraud and into the future of autonomous cars. Although I talk about specific enterprises here, the use cases are pretty generic and horizontal. The fraud detection use case, for example, appeals to a large number of verticals where financial transactions and/or user behavior monitoring is essential, including eCommerce, financial and retail environments.