For example, suppose we want to use a RNN to predict the weather, one hour in advance (based on the weather at say the previous 100 hours as input). If we were to use the output method, at each hour we would need to feed in the full 100 hours of data to predict the weather for hour 101. Then to predict the weather for hour 102, we would need to feed in the full 100 (or 101) hours of data; and so on for hours 103 . Alternatively, we could use the rnnTimeStep method.

Welcome to part 2 of our 4-part blog series. If you want to get started right now, download the complete Deep Learning and Artificial Intelligence white paper. In many contexts, artificial intelligence, machine learning, and deep learning are used interchangeably, but in reality, machine and deep learning are subsets of AI. We can think of AI as the branch of computer science focused on building machines capable of intelligent behavior, while machine learning and deep learning are practices of using algorithms to sift through data, learn from the data, and make predictions or take autonomous actions. Therefore, instead of programming specific constraints for an algorithm to follow, the algorithm is trained using large amounts of data to give it the ability to independently learn, reason, and perform a specific task.

I have spoken to about 150 VPs of Marketing and CMOs from around the country, and have gotten more feedback in writing with the LinkedIn messenger. The questions I'm asking are very general and sometimes hard to answer concisely, but it gives us a good feel for how versed marketers are with what Artificial Intelligence (AI) means, exactly, and what the differences are between AI and Machine Learning and Deep Learning. My first impression is that marketers are very up to speed on what these terms mean then I would have originally guessed. Most marketers (from all age groups I might add) state that we don't really have "true" Artificial Intelligence. We are still a few years away from true AI (actually defining true AI is a topic for another post).

In my last article, I discussed the fundamentals of deep learning, where I explained the basic working of a artificial neural network. If you've been following this series, today we'll become familiar with practical process of implementing neural network in Python (using Theano package). I found various other packages also such as Caffe, Torch, TensorFlow etc to do this job. But, Theano is no less than and satisfactorily execute all the tasks. Also, it has multiple benefits which further enhances the coding experience in Python. In this article, I'll provide a comprehensive practical guide to implement Neural Networks using Theano.

As we move into 2018, the monthly Datasets Publishing Awards has concluded. We're pleased to have recognized many publishers of high-quality, original, and impactful datasets. It was only a little over a year ago that we opened up our public Datasets platform to data enthusiasts all over the world to share their work. We've now reached almost 10,000 public datasets, making choosing winners each month a difficult task! These interviews feature the stories and backgrounds of the November and December winners of the prize.

In the current deep learning era, there is a neural network architecture available for possibly every problem; for instance, ResNet / CNN for computer vision problems, RNN with attention for Language / Speech problems and so on. The Deep Learning models have surpassed almost all the traditional Machine Learning techniques and are the current SOTA for various problems that were deemed to be impossible to achieve for a computer. However, if we take a deep look at the primary computations performed by a neural network, we can obtain better insights about the intricate details of its working. One prominent phenomenon that I observed is that the norm of the data that we process through the network changes to a certain extent. In order to address this problem, the use of batch-normalization layer is common.

Microsoft started its research presence in Montreal a year ago, when it acquired the deep learning startup Maluuba. Microsoft today confirmed its plans to significantly expand its Montreal research lab. They are planning to double the size of lab within the next two years, to as many as 75 technical experts. In addition, Microsoft has hired a renowned AI expert, Geoffrey Gordon as the Montreal research lab's new research director. Jennifer Chayes, a technical fellow and managing director of Microsoft Research New England, New York City and Montreal said that Montreal is really one of the most exciting places in AI right now and Geoffrey Gordon was an obvious choice because of his interest in both the foundational AI research that addresses fundamental AI challenges and the applied work that can quickly find its way into mainstream use.

This nascent AI technique – which requires no input data, substantially less computing power, and in which the evolutionary-like AI learns from itself – could soon evolve to enable its application to real-world problems in the natural sciences. Collaboration with Earth scientists to identify the systems – from climate science, materials science, biology, and other areas – which can be codified to apply reinforcement learning for scientific progress and discovery is vital. For example, DeepMind co-founder, Demis Hassabis, has suggested that in materials science, a descendant of AlphaGo Zero could be used to search for a room temperature superconductor – a hypothetical substance that allows for incredibly efficient energy systems.

In this transcript from an interview conducted by insideHPC, Mike Bernhardt discusses the CANDLE project for cancer research with Rick Stevens from Argonne National Lab. The CANcer Distributed Learning Environment (CANDLE) is an ECP application development project targeting new computational methods for cancer treatment with precision medicine. What is CANDLE all about? It has to do with building a scalable deep-learning environment that can be applied to a variety of problems in cancer, initially. CANDLE is designed to run on the big machines that we have at the US Department of Energy (DOE). The goal is to have an easy-to-use environment that can take advantage of the full power of these big systems to search through large combinations of deep-learning models to find optimal models for making predictions in cancer.

Last year, the best paper award for ICLR 2017 went to "Re-thinking Generalization" by Chiyuan Zhang et al. The key take away of his teams discovery is that the nature of Deep Learning systems is remarkably very different from other classical machine learning systems. One of the biggest misunderstanding about Deep Learning is that it is just a higher dimensional form of curve fitting and thus solved from the perspective of optimization techniques. This is incorrect notion can be due to the fact that the way Artificial Neural Networks (ANN) is taught to many is that it is just a larger form of logistic regression. Alternatively, for the more experienced machine learning expert, everything can be framed from the viewpoint of an optimization problem.