Artificial intelligence (AI) is already besting human intelligence in a number ways. Google and DeepMind's AlphaGo Zero is arguably the greatest Go player in the world, and it learned the game by teaching itself. DeepMind researchers claim they never even reached the limit of the AI's potential, meaning it could be capable of even more impressive tasks. AlphaGoZero is just one of many AIs under development across the globe, and as the industry continues to grow, these systems are going to get smarter and smarter. According to futurist Ian Pearson, humanity's only option if it wants to maintain pace is to merge with AI.

The field of medicine has, arguably, been more positively affected by modern deep learning techniques than any other industry. And, despite the unending deluge of panic-ridden articles declaring AI the path to apocalypse, we're now living in a world where algorithms save lives every day. Welcome to ...

Thinking about depression treatment is likely to depress you. The disease, which is the world's leading cause of disability, affects an estimated 300 million people globally. Although there are plenty of possible treatments, doctors have no reliable way to know what's best for each patient. Many people struggle with depression for years -- enduring medication side effects and feelings of despair -- while they contend with what's now a trial-and-error effort. "That suffering motivated us to find a better way," said Robert Fratila, co-founder and chief technology officer at Montréal-based startup, Aifred Health.

Noah Goodman, a fellow at Uber AI Labs and an associate professor of psychology, computer science, and linguistics at Stanford University, discussed some of the often overlooked but very important "ingredients" in modern machine learning: deep learning classification systems.

The prediction of the gas production from mature gas wells, due to their complex end-of-life behavior, is challenging and crucial for operational decision making. In this paper, we apply a modified deep LSTM model for prediction of the gas flow rates in mature gas wells, including the uncertainties in input parameters. Additionally, due to changes in the system in time and in order to increase the accuracy and robustness of the prediction, the Ensemble Kalman Filter (EnKF) is used to update the flow rate predictions based on new observations. The developed approach was tested on the data from two mature gas production wells in which their production is highly dynamic and suffering from salt deposition. The results show that the flow predictions using the EnKF updated model leads to better Jeffreys' J-divergences than the predictions without the EnKF model updating scheme.

Deep learning methods are useful for high-dimensional data and are becoming widely used in many areas of software engineering. Deep learners utilizes extensive computational power and can take a long time to train-- making it difficult to widely validate and repeat and improve their results. Further, they are not the best solution in all domains. For example, recent results show that for finding related Stack Overflow posts, a tuned SVM performs similarly to a deep learner, but is significantly faster to train. This paper extends that recent result by clustering the dataset, then tuning very learners within each cluster. This approach is over 500 times faster than deep learning (and over 900 times faster if we use all the cores on a standard laptop computer). Significantly, this faster approach generates classifiers nearly as good (within 2\% F1 Score) as the much slower deep learning method. Hence we recommend this faster methods since it is much easier to reproduce and utilizes far fewer CPU resources. More generally, we recommend that before researchers release research results, that they compare their supposedly sophisticated methods against simpler alternatives (e.g applying simpler learners to build local models).

Patients in the intensive care unit (ICU) require constant and close supervision. To assist clinical staff in this task, hospitals use monitoring systems that trigger audiovisual alarms if their algorithms indicate that a patient's condition may be worsening. However, current monitoring systems are extremely sensitive to movement artefacts and technical errors. As a result, they typically trigger hundreds to thousands of false alarms per patient per day - drowning the important alarms in noise and adding to the exhaustion of clinical staff. In this setting, data is abundantly available, but obtaining trustworthy annotations by experts is laborious and expensive. We frame the problem of false alarm reduction from multivariate time series as a machine-learning task and address it with a novel multitask network architecture that utilises distant supervision through multiple related auxiliary tasks in order to reduce the number of expensive labels required for training. We show that our approach leads to significant improvements over several state-of-the-art baselines on real-world ICU data and provide new insights on the importance of task selection and architectural choices in distantly supervised multitask learning.

We introduce Imagination-Augmented Agents (I2As), a novel architecture for deep reinforcement learning combining model-free and model-based aspects. In contrast to most existing model-based reinforcement learning and planning methods, which prescribe how a model should be used to arrive at a policy, I2As learn to interpret predictions from a learned environment model to construct implicit plans in arbitrary ways, by using the predictions as additional context in deep policy networks. I2As show improved data efficiency, performance, and robustness to model misspecification compared to several baselines.

Deep convolutional neural networks have achieved great success in various applications. However, training an effective DNN model for a specific task is rather challenging because it requires a prior knowledge or experience to design the network architecture, repeated trial-and-error process to tune the parameters, and a large set of labeled data to train the model. In this paper, we propose to overcome these challenges by actively adapting a pre-trained model to a new task with less labeled examples. Specifically, the pre-trained model is iteratively fine tuned based on the most useful examples. The examples are actively selected based on a novel criterion, which jointly estimates the potential contribution of an instance on optimizing the feature representation as well as improving the classification model for the target task. On one hand, the pre-trained model brings plentiful information from its original task, avoiding redesign of the network architecture or training from scratch; and on the other hand, the labeling cost can be significantly reduced by active label querying. Experiments on multiple datasets and different pre-trained models demonstrate that the proposed approach can achieve cost-effective training of DNNs.

The standard L-BFGS method relies on gradient approximations that are not dominated by noise, so that search directions are descent directions, the line search is reliable, and quasi-Newton updating yields useful quadratic models of the objective function. All of this appears to call for a full batch approach, but since small batch sizes give rise to faster algorithms with better generalization properties, L-BFGS is currently not considered an algorithm of choice for large-scale machine learning applications. One need not, however, choose between the two extremes represented by the full batch or highly stochastic regimes, and may instead follow a progressive batching approach in which the sample size increases during the course of the optimization. In this paper, we present a new version of the L-BFGS algorithm that combines three basic components - progressive batching, a stochastic line search, and stable quasi-Newton updating - and that performs well on training logistic regression and deep neural networks. We provide supporting convergence theory for the method.