Qualcomm has unveiled its roadmap for bringing artificial intelligence capabilities to smart devices and goals to bring complementary AI features to the cloud. Sparked in 2007, Qualcomm's efforts to improve AI services use neuron-based approaches to machine learning, and their efforts aim to make AI a cornerstone of most digital-enabled products, including automobiles and machinery. Qualcomm's initial AI efforts, before the smart device boom was in full swing, initially focused on motion control and computer vision applications, fields inspired by biological counterparts. Their efforts later extended into neural net fields supplemented by deep learning algorithms. On August 16, 2017, Qualcomm further expanded their capabilities by purchasing the University of Amsterdam-affiliated Scyfer BV, an AI-focused company with experience in healthcare, finance, and manufacturing.

Scyfer was established about four years ago at Amsterdam Science Park as a spinoff of the University of Amsterdam. The company works with the latest and most advanced technologies in the field of deep learning. Founder and CEO of Scyfer is Max Welling, professor of Machine Learning at the Faculty of Natural Sciences, Mathematics and Informatics at the University of Amsterdam. He is also employed at the University of California Irvine, the Canadian Institute for Advanced Research (CIFAR).

Henri Waelbroeck, director of research at machine learning trade execution system Portware, says rather poetically that the system "reads the tea leaves" in market data to distinguish different sorts of orders and execute trades more efficiently. Portware uses artificial intelligence to help traders select the best algorithm for particular market conditions, asset class, broker, venue etc., interacting with the order flow and computing a mind-boggling array of variables in real time. Say you are buying a stock, and you predict there is likely to be more orders hitting the bid side of the spread in the next five minutes, you should be able to operate an efficient algorithm that only posts limit orders and collects the spread as it executes. Using an algorithm that crosses the spread in this instance would be wasteful since you expect order flow to be coming your way. Waelbroeck, formerly a professor at the Institute of Nuclear Sciences at the National University of Mexico, whose specialisms include genetic algorithms and chaos theory, said: "Just throwing machine learning at problems usually doesn't give a very good answer. You need to have a good analytical understanding of what's going on and this usually gives you a baseline model and then you find opportunities to insert machine learning tactically to exploit opportunities to improve the models."

TensorFlow enables developers to quickly and easily get started with deep learning in the cloud. The framework has broad support in the industry and has become a popular choice for deep learning research and application development, particularly in areas such as computer vision, natural language understanding and speech translation. You can get started using TensorFlow on AWS by launching the AWS Deep Learning AMI which comes bundled with TensorFlow; as well as, other popular deep learning frameworks such as Apache MXNet, Caffe, Caffe2, Theano, Torch, Keras, and the Microsoft Cognitive Toolkit.

Our Dota 2 result shows that self-play can catapult the performance of machine learning systems from far below human level to superhuman, given sufficient compute. In the span of a month, our system went from barely matching a high-ranked player to beating the top pros and has continued to improve since then. Supervised deep learning systems can only be as good as their training datasets, but in self-play systems, the available data improves automatically as the agent gets better. Improvements came from every part of the system, from adding new features to algorithmic improvements to scaling things up. The graph is surprisingly linear, meaning the team improved the bot exponentially over time.

SITTING IN AN office in San Francisco, Igor Barani calls up some medical scans on his screen. He is the chief executive of Enlitic, one of a host of startups applying deep learning to medicine, starting with the analysis of images such as X-rays and CT scans. It is an obvious use of the technology. Deep learning is renowned for its superhuman prowess at certain forms of image recognition; there are large sets of labelled training data to crunch; and there is tremendous potential to make health care more accurate and efficient. Dr Barani (who used to be an oncologist) points to some CT scans of a patient's lungs, taken from three different angles.

Over the past few months, I have been collecting AI cheat sheets. From time to time I share them with friends and colleagues and recently I have been getting asked a lot, so I decided to organize and share the entire collection. To make things more interesting and give context, I added descriptions and/or excerpts for each major topic.

For those who don't know, the Enigma machine was most famously used by the Germans during World War II to encrypt and decrypt messages. Give the neural network some encrypted text, called the ciphertext, along with the three-letter key that was used to encrypt the text, and the network predicts what the original text, or plaintext, was with around 96-97% accuracy. The type of neural network he used was a Long Short Term Memory (LSTM) network, a type of Recurrent Neural Network (RNN) that we talked about in our article covering many of the different types of neural networks developed over the years. RNNs are Turing-complete, meaning they can approximate any function. How did [Sam] do it?

Raman spectroscopy is a ubiquitous method for characterisation of substances in a wide range of settings including industrial process control, planetary exploration, homeland security, life sciences, geological field expeditions and laboratory materials research. In all of these environments there is a requirement to identify substances from their Raman spectrum at high rates and often in high volumes. Whilst machine classification has been demonstrated to be an essential approach to achieve real time identification, it still requires preprocessing of the data. This is true regardless of whether peak detection or multivariate methods, operating on whole spectra, are used as input. A standard pipeline for a machine classification system based on Raman spectroscopy includes preprocessing in the following order: cosmic ray removal, smoothing and baseline correction.

The analysis of mixed data has been raising challenges in statistics and machine learning. One of two most prominent challenges is to develop new statistical techniques and methodologies to effectively handle mixed data by making the data less heterogeneous with minimum loss of information. The other challenge is that such methods must be able to apply in large-scale tasks when dealing with huge amount of mixed data. To tackle these challenges, we introduce parameter sharing and balancing extensions to our recent model, the mixed-variate restricted Boltzmann machine (MV.RBM) which can transform heterogeneous data into homogeneous representation. We also integrate structured sparsity and distance metric learning into RBM-based models. Our proposed methods are applied in various applications including latent patient profile modelling in medical data analysis and representation learning for image retrieval. The experimental results demonstrate the models perform better than baseline methods in medical data and outperform state-of-the-art rivals in image dataset.