University of Alberta scientists developed a deep learning-based prostate cancer diagnostic platform that only uses a single drop of blood which will allow men to bypass the current painful biopsy methods. Using a GTX 1060 GPU, CUDA and the MathWorks Neural Network Toolbox, the scientists' trained their model on information from millions of cancer cell nanoparticles in the blood to recognize the unique fingerprint of aggressive prostate cancer. To test their method, they evaluated a group of 377 men who were referred to their urologist with suspected prostate cancer and found that their system called Extracellular Vesicle Fingerprint Predictive Score (EV-FPS) correctly identified men with aggressive prostate cancer 40 percent more accurately than the most common test in wide use today. "Higher sensitivity means that our test will miss fewer aggressive cancers," said John Lewis, the Alberta Cancer Foundation's Frank and Carla Sojonky Chair of Prostate Cancer Research at the University of Alberta. "For this kind of test you want the sensitivity to be as high as possible because you don't want to miss a single cancer that should be treated."

Machine learning and deep learning can be powerful toolkits for enterprise CIOs – but how can you tell which framework is best for your business? The prospect of deep learning doing all the heavy lifting when it comes to building a prediction model sounds very exciting. However, it poses significant challenges when it comes to organizational problem solving due to the following reasons. Cloud-based compute is usually the most feasible option, but may not sit well with companies that are very guarded about their data going off premise. On the other hand, on-premise servers and staff to run and maintain them are very costly.

Azure Machine Learning is an integrated, end-to-end data science and advanced analytics solution. It enables data scientists to prepare data, develop experiments, and deploy models at cloud scale. Together, these applications and services help significantly accelerate your data science project development and deployment. Azure Machine Learning fully supports open source technologies. You can execute your experiments in managed environments such as Docker containers and Spark clusters.

The Compunet Research Line at Istituto Italiano di Tecnologia (IIT) in Genoa is opening a postdoctoral position (2 years) with focus on bioinformatics and machine learning: in particular deep learning for genomics for disease/pharmaceutical predictions. Knowledge of computational drug discovery will be considered a plus. Knowledge of Caffe, Theano, TensorFlow or related deep learning GPU enabled/ machine learning software libraries is also a desirable skill. The salary will be internationally competitive and commensurate with the candidate's experience. Applicants are invited to submit a cover letter stating research goals, a curriculum vitae and contact information of 2 referees to panagiota.dimopoulou@iit.it

AI deep learning expert and University of Montreal Professor Yoshua Bengio talks about deep learning--what it is, how it got there, where it's going, and how you can learn more about it.

I don't know who produced this image, but I've found it posted on the Deep Learning group on Facebook. For other data science concepts explained in one picture, click here.

Deep neural networks continue to show improved performance with increasing depth, an encouraging trend that implies an explosion in the possible permutations of network architectures and hyperparameters for which there is little intuitive guidance. To address this increasing complexity, we propose Evolutionary DEep Networks (EDEN), a computationally efficient neuro-evolutionary algorithm which interfaces to any deep neural network platform, such as TensorFlow. We show that EDEN evolves simple yet successful architectures built from embedding, 1D and 2D convolutional, max pooling and fully connected layers along with their hyperparameters. Evaluation of EDEN across seven image and sentiment classification datasets shows that it reliably finds good networks -- and in three cases achieves state-of-the-art results -- even on a single GPU, in just 6-24 hours. Our study provides a first attempt at applying neuro-evolution to the creation of 1D convolutional networks for sentiment analysis including the optimisation of the embedding layer.

Deep neural networks (NN) are extensively used for machine learning tasks such as image classification, perception and control of autonomous systems. Increasingly, these deep NNs are also been deployed in high-assurance applications. Thus, there is a pressing need for developing techniques to verify neural networks to check whether certain user-expected properties are satisfied. In this paper, we study a specific verification problem of computing a guaranteed range for the output of a deep neural network given a set of inputs represented as a convex polyhedron. Range estimation is a key primitive for verifying deep NNs. We present an efficient range estimation algorithm that uses a combination of local search and linear programming problems to efficiently find the maximum and minimum values taken by the outputs of the NN over the given input set. In contrast to recently proposed "monolithic" optimization approaches, we use local gradient descent to repeatedly find and eliminate local minima of the function. The final global optimum is certified using a mixed integer programming instance. We implement our approach and compare it with Reluplex, a recently proposed solver for deep neural networks. We demonstrate the effectiveness of the proposed approach for verification of NNs used in automated control as well as those used in classification.

Auto-encoding is an important task which is typically realized by deep neural networks (DNNs) such as convolutional neural networks (CNN). In this paper, we propose EncoderForest (abbrv. eForest), the first tree ensemble based auto-encoder. We present a procedure for enabling forests to do backward reconstruction by utilizing the equivalent classes defined by decision paths of the trees, and demonstrate its usage in both supervised and unsupervised setting. Experiments show that, compared with DNN autoencoders, eForest is able to obtain lower reconstruction error with fast training speed, while the model itself is reusable and damage-tolerable.

This paper proposes a minimal contrastive divergence method for learning energy-based generative ConvNet models of images at multiple grids (or scales) simultaneously. For each grid, we learn an energy-based probabilistic model where the energy function is defined by a bottom-up convolutional neural network (ConvNet or CNN). Learning such a model requires generating synthesized examples from the model. Within each iteration of our learning algorithm, for each observed training image, we generate synthesized images at multiple grids by initializing the finite-step MCMC sampling from a minimal 1 x 1 version of the training image. The synthesized image at each subsequent grid is obtained by a finite-step MCMC initialized from the synthesized image generated at the previous coarser grid. After obtaining the synthesized examples, the parameters of the models at multiple grids are updated separately and simultaneously based on the differences between synthesized and observed examples. We call this learning method the multi-grid minimal contrastive divergence. We show that this method can learn realistic energy-based generative ConvNet models, and it outperforms the original contrastive divergence (CD) and persistent CD.