Analyzing large-scale, multi-experiment studies requires scientists to test each experimental outcome for statistical significance and then assess the results as a whole. We present Black Box FDR (BB-FDR), an empirical-Bayes method for analyzing multi-experiment studies when many covariates are gathered per experiment. BB-FDR learns a series of black box predictive models to boost power and control the false discovery rate (FDR) at two stages of study analysis. In Stage 1, it uses a deep neural network prior to report which experiments yielded significant outcomes. In Stage 2, a separate black box model of each covariate is used to select features that have significant predictive power across all experiments. In benchmarks, BB-FDR outperforms competing state-of-the-art methods in both stages of analysis. We apply BB-FDR to two real studies on cancer drug efficacy. For both studies, BB-FDR increases the proportion of significant outcomes discovered and selects variables that reveal key genomic drivers of drug sensitivity and resistance in cancer.

As Machine Learning (ML) gets applied to security-critical or sensitive domains, there is a growing need for integrity and privacy guarantees for ML computations running in untrusted environments. A pragmatic solution comes from Trusted Execution Environments, which use hardware and software protections to isolate sensitive computations from the untrusted software stack. However, these isolation guarantees come at a price in performance, compared to untrusted alternatives. This paper initiates the study of high performance execution of Deep Neural Networks (DNNs) in trusted environments by efficiently partitioning computations between trusted and untrusted devices. Building upon a simple secure outsourcing scheme for matrix multiplication, we propose Slalom, a framework that outsources execution of all linear layers in a DNN from any trusted environment (e.g., SGX, TrustZone or Sanctum) to a faster co-located device. We evaluate Slalom by executing DNNs in an Intel SGX enclave, which selectively outsources work to an untrusted GPU. For two canonical DNNs, VGG16 and MobileNet, we obtain 20x and 6x increases in throughput for verifiable inference, and 10x and 3.5x for verifiable and private inference.

One motivation for learning generative models of environments is to use them as simulators for model-based reinforcement learning. Yet, it is intuitively clear that when time horizons are long, rolling out single step transitions is inefficient and often prohibitive. In this paper, we propose a generative model that learns state representations containing explicit beliefs about states several time steps in the future and that can be rolled out directly in these states without executing single step transitions. The model is trained on pairs of temporally separated time points, using an analogue of temporal difference learning used in reinforcement learning, taking the belief about possible futures at one time point as a bootstrap for training the belief at an earlier time. While we focus purely on the study of the model rather than its use in reinforcement learning, the model architecture we design respects agents' constraints as it builds the representation online.

Developing an autonomous vehicle requires a massive amount of data. Before any AV can safely navigate on the road, engineers must first train the artificial intelligence (AI) algorithms that enable the car to drive itself. Deep learning, a form of AI, is used to perceive the environment surrounding the car and to make driving decisions with superhuman levels of performance and precision. This is an enormous big data challenge. A single test vehicle can generate petabytes of data a year.

Word2Vec model is used for learning vector representations of words called "word embeddings". This is typically done as a preprocessing step, after which the learned vectors are fed into a discriminative model (typically an RNN) to generate predictions and perform all sort of interesting things. Image and audio processing systems work with rich, high-dimensional datasets encoded as vectors of the individual raw pixel-intensities for image data, thus all the information is encoded in the data hence the relation between various entities in the system like (cats and dogs) can be established. But, when it comes to natural language processing systems traditionally it treats words as discrete atomic symbols, and therefore'cat' may be represented as Id537 and'dog' as Id143.These encodings are arbitrary, and provide no useful information to the system regarding the relationships that may exist between the individual symbols. This means that the model can leverage very little of what it has learned about'cats' when it is processing data about'dogs' (such that they are both animals, four-legged, pets, etc.).

It's no exaggeration to call Element AI one of the top startups in the world right now. With the help of deep learning pioneer Yoshua Bengio, the company is making AI-powered products for the enterprise. And from its beginning in October 2016, Element AI has broken the rules of what to expect from a startup. In December 2016, Element AI was the very first company to receive funding from Microsoft Ventures. Six months later, the company raised a $102 million series A round.

Artificial intelligence refers to the ability of computers systems to perform tasks that normally require human intellect and judgment. Also called machine learning, these systems take in data that is structured or unstructured producing algorithms, finding patterns, and continually learning and refining its capacity over time to make decisions, guide behavior, offer solutions, or act. AI is now moving rapidly into "deep learning" which looks not just at the face value of data but draws meaning and representations from initial and cumulative data accelerating its learning and capacity as it grows. IBM's Watson is probably the most commonly know system exercising deep learning. Although the term was first coined in the mid 1950s, we are beginning to see their use in everyday life with self-driving cars, airplane autopilot systems, automatically controlling traffic lights based on flow and density, mobile check deposits deciphering handwriting, detecting human emotion to target advertising, and analyzing electronic trails to learn and exploit human behavior patterns.

And you will be on the Jupyter Notebook with all the installed packages:). A note here: The Spark Notebook (DLS SPARK) is an upcoming feature which will be released to public sometime next month and tell that it is still in private beta (just for this post). The first step to applying deep learning on images is the ability to load the images. Deep Learning Pipelines includes utility functions that can load millions of images into a DataFrame and decode them automatically in a distributed fashion, allowing manipulation at scale. The new version of spark (2.3.0) has this ability too but we will be using the sparkdl library.

Machine learning, a collection of data-analytical techniques aimed at building predictive models from multi-dimensional datasets, is becoming integral to modern biological research. By enabling one to generate models that learn from large datasets and make predictions on likely outcomes, machine learning can be used to study complex cellular systems such as biological networks. Here, we provide a primer on machine learning for life scientists, including an introduction to deep learning. We discuss opportunities and challenges at the intersection of machine learning and network biology, which could impact disease biology, drug discovery, microbiome research, and synthetic biology.

Inception Network Deep Learning Architecture Sharing some of my experience on simplifying the cost & how to go with best combination of hyper parameters.