This paper presents a deep learning method for faster magnetic resonance imaging (MRI) by reducing k-space data with sub-Nyquist sampling strategies and provides a rationale for why the proposed approach works well. Uniform subsampling is used in the time-consuming phase-encoding direction to capture high-resolution image information, while permitting the image-folding problem dictated by the Poisson summation formula. To deal with the localization uncertainty due to image folding, very few low-frequency k-space data are added. Training the deep learning net involves input and output images that are pairs of Fourier transforms of the subsampled and fully sampled k-space data. Numerous experiments show the remarkable performance of the proposed method; only 29% of k-space data can generate images of high quality as effectively as standard MRI reconstruction with fully sampled data.

We propose a new algorithm called Parle for parallel training of deep networks that converges 2-4x faster than a data-parallel implementation of SGD, while achieving significantly improved error rates that are nearly state-of-the-art on several benchmarks including CIFAR-10 and CIFAR-100, without introducing any additional hyper-parameters. We exploit the phenomenon of flat minima that has been shown to lead to improved generalization error for deep networks. Parle requires very infrequent communication with the parameter server and instead performs more computation on each client, which makes it well-suited to both single-machine, multi-GPU settings and distributed implementations.

Machine learning is transforming business. But even as the technology advances, companies still struggle to take advantage of it, largely because they don't understand how to strategically implement machine learning in service of business goals. Hype hasn't helped, sowing confusion over what exactly machine learning is, how well it works and what it can do for your company. Here, we provide a clear-eyed look at what machine learning is and how it can be used today. Machine learning is a subset of artificial intelligence that enables systems to learn and predict outcomes without explicit programming.

In case you didn't make it to the ISCA (International Society for Computers and their Applications) session this year, you might be interested in a presentation by [Joel Emer] an MIT professor and scientist for NVIDIA. Along with another MIT professor and two PhD students ([Vivienne Sze], [Yu-Hsin Chen], and [Tien-Ju Yang]), [Emer's] presentation covers hardware architectures for deep neural networks. The presentation covers the background on deep neural networks and basic theory. Then it progresses to deep learning specifics. One interesting graph shows how neural networks are getting better at identifying objects in images every year and as of 2015 can do a better job than a human over a set of test images.

The authors of this paper propose a method to increase training speed by freezing layers. The authors demonstrated a way to freeze the layers one by one as soon as possible, resulting in fewer and fewer backward passes, which in turn lowers training time. The authors experimented with different values for Equation 2.1 The authors tried scaling the initial learning rate so that each layer was trained for an equal amount of time. I demonstrated 2(and half of my own) very recent and novel techniques to improve accuracy and lower training time by fine tuning learning rates.

Accurate time series forecasting during high variance segments (e.g., holidays and sporting events) is critical for anomaly detection, resource allocation, budget planning, and other related tasks necessary to facilitate optimal Uber user experiences at scale. Forecasting these variables, however, can be challenging because extreme event prediction depends on weather, city population growth, and other external factors that contribute to forecast uncertainty. In recent years, the Long Short Term Memory (LSTM) technique has become a popular time series modeling framework due to its end-to-end modeling, ease of incorporating exogenous variables, and automatic feature extraction abilities. Uncertainty estimation in deep learning remains a less trodden but increasingly important component of assessing forecast prediction truth in LSTM models. Through our research, we found that a neural network forecasting model is able to outperform classical time series methods in use cases with long, interdependent time series. While beneficial in other ways, our new model did not offer insights into prediction uncertainty, which helps determine how much we can trust the forecast.

This course provides you to be able to build Deep Neural Networks models for different business domains with one of the most common machine learning library TensorFlow provided by Google AI team. The both concept of deep learning and its applications will be mentioned in this course. Also, we will focus on Keras. Also, you don't have to be attend any ML course before.

When the science of IBM Research comes together with a Cloud service engineering team, something special happens. That is the case for the Watson Visual Recognition cloud service delivered by my team. In December, we launched a significant update to our Deep Learning based Watson Visual Recognition service for how Watson can see and identify subject matter of images. A recent blog post by Recast AI has provided a good summary of this marketplace and the remarkable performance that is being achieved by the Watson Visual Recognition service. When you bring together Watson Visual Recognition's market-leading performance with your ability to extend this cloud-based service with custom visual recognition classifiers, the solutions you can build are boundless.

Time-series data arise in many fields including finance, signal processing, speech recognition and medicine. A standard approach to time-series problems usually requires manual engineering of features which can then be fed into a machine learning algorithm. Engineering of features generally requires some domain knowledge of the discipline where the data has originated from. For example, if one is dealing with signals (i.e. A similar situation arises in image classification, where manually engineered features (obtained by applying a number of filters) could be used in classification algorithms.

Valentino Zocca graduated with a PhD in mathematics from the University of Maryland, USA, with a dissertation in symplectic geometry, after having graduated with a laurea in mathematics from the University of Rome. He spent a semester at the University of Warwick. After a post-doc in Paris, Valentino started working on hightech projects in the Washington, D.C. area and played a central role in the design, development, and realization of an advanced stereo 3D Earth visualization software with head tracking at Autometric, a company later bought by Boeing. At Boeing, he developed many mathematical algorithms and predictive models, and using Hadoop, he has also automated several satellite-imagery visualization programs. He has since become an expert on machine learning and deep learning and has worked at the U.S. Census Bureau and as an independent consultant both in the US and in Italy.