AWNBench is a Stanford University project designed to allow different deep learning methods to be compared by running a number of competitions. Two parts of the Dawnbench competition attracted our attention, the CIFAR 10 and Imagenet competitions. Their goal was simply to deliver the fastest image classifier as well as the cheapest one to achieve a certain accuracy (93% for Imagenet, 94% for CIFAR 10). In the CIFAR 10 competition our entries won both training sections: fastest, and cheapest. In this post we'll discuss our approach to each competition.

The future depends on connectivity. From artificial intelligence and self-driving cars to telemedicine and mixed reality to as yet undreamt technologies, all the things we hope will make our lives easier, safer, and healthier will require high-speed, always-on internet connections. To keep up with the explosion of new connected gadgets and vehicles, not to mention the deluge of streaming video, the mobile industry is working on something called 5G--so named because it's the fifth generation of wireless networking technology. The promise is that 5G will bring speeds of around 10 gigabits per second to your phone. US carriers promise that 5G will be available nationwide by 2020, but the first 5G networks won't be nearly so fast. Carriers have launched demos and pilot programs that demonstrate big leaps in wireless performance, but mobile networks based on the "millimeter-wave" technology that may deliver the fastest speeds probably won't be widely available for years.

No one likes it when a stranger butts into their conversation. Especially when they interrupt with some astonishing non-sequitur. Create a free Amazon Business account and get a 30-day free trial to Business Prime. You're watching TV and chatting about the painfully demanding couple on House Hunters International when a distant voice pipes up: "The circumference of the Earth is 24,901 miles." Or whatever you call your Google Home person.

Hyderabad: The future is going to be heavily dependent on Artificial Intelligence (AI) and more than 1,000 companies in India claim to be working on AI in some form, according to Suman Reddy, Managing Director, Pegasystems India. Speaking to Telangana Today on emerging technologies, Reddy said major industries in banking, government and automobile manufacturing would use AI to stay competitive while everything would gradually be powered by AI, right from televisions to home appliances. "Everyone from consumer internet, technology, e-commerce, finance, healthcare companies to financial services and automakers are betting big on AI and other new age technologies," he said. With AI expected to fuel almost everything that helps and empowers daily lives, the demand for AI is going to multiply manifold and India will do well to fill this demand and achieve leadership positions by having captive tech talent at the global stage. On the present situation in IT sector, he said one would have to keep an eye on the prevailing industry trends along with constantly testing their skills.

Jefferson Lab is hosting an A.I. Hack-A-Thon for those interested in learning about artificial intelligence. The purpose is to generate interest in A.I. in the field of nuclear physics by giving participants a free, hands on experience, according to the news release. The event is free and open to the public. The deadline to register is Friday. "The last 10 years have seen explosive growth in the field of A.I." according to the news release.

Normalizing flows transform a simple base distribution into a complex target distribution and have proved to be powerful models for data generation and density estimation. In this work, we propose a novel type of normalizing flow driven by a differential deformation of the continuous-time Wiener process. As a result, we obtain a rich time series model whose observable process inherits many of the appealing properties of its base process, such as efficient computation of likelihoods and marginals. Furthermore, our continuous treatment provides a natural framework for irregular time series with an independent arrival process, including straightforward interpolation. We illustrate the desirable properties of the proposed model on popular stochastic processes and demonstrate its superior flexibility to variational RNN and latent ODE baselines in a series of experiments on synthetic and real-world data.

Accurate predictions of reactive mixing are critical for many Earth and environmental science problems. To investigate mixing dynamics over time under different scenarios, a high-fidelity, finite-element-based numerical model is built to solve the fast, irreversible bimolecular reaction-diffusion equations to simulate a range of reactive-mixing scenarios. A total of 2,315 simulations are performed using different sets of model input parameters comprising various spatial scales of vortex structures in the velocity field, time-scales associated with velocity oscillations, the perturbation parameter for the vortex-based velocity, anisotropic dispersion contrast, and molecular diffusion. Outputs comprise concentration profiles of the reactants and products. The inputs and outputs of these simulations are concatenated into feature and label matrices, respectively, to train 20 different machine learning (ML) emulators to approximate system behavior. The 20 ML emulators based on linear methods, Bayesian methods, ensemble learning methods, and multilayer perceptron (MLP), are compared to assess these models. The ML emulators are specifically trained to classify the state of mixing and predict three quantities of interest (QoIs) characterizing species production, decay, and degree of mixing. Linear classifiers and regressors fail to reproduce the QoIs; however, ensemble methods (classifiers and regressors) and the MLP accurately classify the state of reactive mixing and the QoIs. Among ensemble methods, random forest and decision-tree-based AdaBoost faithfully predict the QoIs. At run time, trained ML emulators are $\approx10^5$ times faster than the high-fidelity numerical simulations. Speed and accuracy of the ensemble and MLP models facilitate uncertainty quantification, which usually requires 1,000s of model run, to estimate the uncertainty bounds on the QoIs.

Encoding domain knowledge into the prior over the high-dimensional weight space is challenging in Bayesian neural networks. Two types of domain knowledge are commonly available in scientific applications: 1. feature sparsity (number of relevant features); 2. signal-to-noise ratio, quantified, for instance, as the proportion of variance explained (PVE). We show both types of domain knowledge can be encoded into the widely used Gaussian scale mixture priors with Automatic Relevance Determination. Specifically, we propose a new joint prior over the local (i.e., feature-specific) scale parameters to encode the knowledge about feature sparsity, and an algorithm to determine the global scale parameter (shared by all features) according to the PVE. Empirically, we show that the proposed informative prior improves prediction accuracy on publicly available datasets and in a genetics application.

Multiple sclerosis is one of the most common chronic neurological diseases affecting the central nervous system. Lesions produced by the MS can be observed through two modalities of magnetic resonance (MR), known as T2W and FLAIR sequences, both providing useful information for formulating a diagnosis. However, long acquisition time makes the acquired MR image vulnerable to motion artifacts. This leads to the need of accelerating the execution of the MR analysis. In this paper, we present a deep learning method that is able to reconstruct subsampled MR images obtained by reducing the k-space data, while maintaining a high image quality that can be used to observe brain lesions. The proposed method exploits the multimodal approach of neural networks and it also focuses on the data acquisition and processing stages to reduce execution time of the MR analysis. Results prove the effectiveness of the proposed method in reconstructing subsampled MR images while saving execution time.

We study the problem of how to efficiently explore zero-sum games whose payoffs and dynamics are initially unknown. The agent is given a certain number of episodes to learn as much useful information about the game as possible. During this learning, the rewards obtained in the game are fictional and thus do not count toward the evaluation of the final strategy. After this exploration phase, the agent must recommend a strategy that should be minimally exploitable by an adversary (who has complete knowledge of the environment and can thus play optimally against it). This setup is called pure exploration in the single-agent reinforcement learning literature. This is an important problem for simulation-based games in which a black-box simulator is queried with strategies to obtain samples of the players' resulting utilities [33], as opposed to the rules of the game being explicitly given. For example, in many military settings, war game simulators are used to generate strategies, and then the strategies need to be ready to deploy in case of actual war [17]. Another prevalent example is finance, where trading strategies are generated in simulation, and then they need to be ready for live trading. A third example is video games such as Dota 2 [4] and Starcraft II [31], where AIs can be trained largely through self-play.