Machine learning is becoming extremely powerful, but it requires extreme amounts of data. You can, for instance, train a deep-learning algorithm to recognize a cat with a cat-fancier's level of expertise, but you'll need to feed it tens or even hundreds of thousands of images of felines, capturing a huge amount of variation in size, shape, texture, lighting, and orientation. It would be lot more efficient if, a bit like a person, an algorithm could develop an idea about what makes a cat a cat from fewer examples. A Boston-based startup called Gamalon has developed technology that lets computers do this in some situations, and it is releasing two products Tuesday based on the approach. If the underlying technique can be applied to many other tasks, then it could have a big impact.

Deep-learning framework Caffe is "made with expression, speed, and modularity in mind." Originally developed in 2013 for machine vision projects, Caffe has since expanded to include other applications, such as speech and multimedia. Speed is a major priority, so Caffe has been written entirely in C, with CUDA acceleration support, although it can switch between CPU and GPU processing as needed. The distribution includes a set of free and open source reference models for common classification jobs, with other models created and donated by the Caffe user community. A new iteration of Caffe backed by Facebook, called Caffe2, is currently under development for a 1.0 release.

Training deep networks is expensive and time-consuming with the training period increasing with data size and growth in model parameters. In this paper, we provide a framework for distributed training of deep networks over a cluster of CPUs in Apache Spark. The framework implements both Data Parallelism and Model Parallelism making it suitable to use for deep networks which require huge training data and model parameters which are too big to fit into the memory of a single machine. It can be scaled easily over a cluster of cheap commodity hardware to attain significant speedup and obtain better results making it quite economical as compared to farm of GPUs and supercomputers. We have proposed a new algorithm for training of deep networks for the case when the network is partitioned across the machines (Model Parallelism) along with detailed cost analysis and proof of convergence of the same. We have developed implementations for Fully-Connected Feedforward Networks, Convolutional Neural Networks, Recurrent Neural Networks and Long Short-Term Memory architectures. We present the results of extensive simulations demonstrating the speedup and accuracy obtained by our framework for different sizes of the data and model parameters with variation in the number of worker cores/partitions; thereby showing that our proposed framework can achieve significant speedup (upto 11X for CNN) and is also quite scalable.

Sepsis is a poorly understood and potentially life-threatening complication that can occur as a result of infection. Early detection and treatment improves patient outcomes, and as such it poses an important challenge in medicine. In this work, we develop a flexible classifier that leverages streaming lab results, vitals, and medications to predict sepsis before it occurs. We model patient clinical time series with multi-output Gaussian processes, maintaining uncertainty about the physiological state of a patient while also imputing missing values. The mean function takes into account the effects of medications administered on the trajectories of the physiological variables. Latent function values from the Gaussian process are then fed into a deep recurrent neural network to classify patient encounters as septic or not, and the overall model is trained end-to-end using back-propagation. We train and validate our model on a large dataset of 18 months of heterogeneous inpatient stays from the Duke University Health System, and develop a new "real-time" validation scheme for simulating the performance of our model as it will actually be used. Our proposed method substantially outperforms clinical baselines, and improves on a previous related model for detecting sepsis. Our model's predictions will be displayed in a real-time analytics dashboard to be used by a sepsis rapid response team to help detect and improve treatment of sepsis.

Co-founder and CEO Nigel Toon laughs at that interview opener -- perhaps because he sold his previous company to the chipmaker back in 2011. "I'm sure Nvidia will be successful as well," he ventures. "They're already being very successful in this market… And being a viable competitor and standing alongside them, I think that would be a worthy aim for ourselves." Toon also flags what he couches an "interesting absence" in the competitive landscape vis-a-vis other major players "that you'd expect to be there" -- e.g. A recent report by analyst Gartner suggests AI technologies will be in almost every software product by 2020.

This article is part of a series. Check out the full series: Part 1, Part 2, Part 3, Part 4, Part 5, Part 6, Part 7 and Part 8! Almost as long as programmers have been writing computer programs, computer hackers have been figuring out ways to exploit those programs. Malicious hackers take advantage of the tiniest bugs in programs to break into systems, steal data and generally wreak havoc. But systems powered by deep learning algorithms should be safe from human interference, right? How is a hacker going to get past a neural network trained on terabytes of data?

My name is Kamal thakur, I am an Electronics Engineer and electronic hobbyist with an interest in making embedded systems, Robotics understandable and enjoyable to other enthusiasts of all experience and knowledge levels. Experienced with project design, development & commissioning, product & application technical support, training & consulting services with international environment. Always eager to learn, I invested a lot of time in learning and teaching, covering a wide range of different scientific topics. Being an electronics engineer, Today I am passionate about data science, artificial intelligence and deep learning for Robotics. I will do my very best to convey my passion for data science to you.

Data is important in healthcare. How a chart is read, if a doctor has time to take a second look at that scan of your chest, and whether there's enough evidence to make you book an appointment with your GP could mean the difference between life and death for you -- and of lower-cost preventative measures and expensive treatments for insurance companies. Currently, we rely on doctors and nurses to interpret key information -- but machines are already coming to their aid, scanning images for signs of cancer, analysing data for symptoms of kidney failure, and more. In the future, apps will allow you to ask Alexa for medical advice, tools will assist GPs with triage and hunt for signs of cancer in medical scans, and chatbots could help treat mental illness. Google-owned DeepMind is working on projects to analyse patient data to predict kidney failure, spot head and neck cancers, and to read complicated eye images.

For example, a major scientific achievement in computing this past year featured reinforcement-learning programs by a computer that outperformed humans in AlphaGo, a massively complex game that build on learning algorithms, modeled after the ancient board game, Go. These DeepMind neural networks function differently than other AI platforms such as IBM's Deep Blue program or Watson, which were assembled from large databases and developed for a pre-defined purpose, and which only function within its scope. Whereas computer programs had previously used Monte Carlo tree search algorithm strategies (search engine programs designed to instruct plays in computer games) to find its moves in the game – like a computer using a database of options programmed by a human in order to select the next proper chess move. However, in these recent advances, the Google Deep Mind programs that operate AlphaGo are able to select moves based on knowledge the programs learned themselves from machine learning in an artificial neural network based on human and computer play. Deep learning, for example with DeepMind, demonstrates that the computational system is not pre-programmed and is in fact a neural network of algorithms that learn from experience, using only raw pixels as data input.

Image recognition is important for many of the advanced technologies we use today. It is used in visual surveillance, guiding autonomous vehicles and even identifying ailments from X-ray images. Most modern smartphones also come with image recognition apps that convert handwriting into typed words. In this chapter we will look at how we can train an ANN algorithm to recognize images of handwritten digits. We will be using the images from the famous MNIST (Mixed National Institute of Standards and Technology) database. Note that this post is written by Kenneth Soo.