Our latest robotics techniques allow robot controllers, trained entirely in simulation and deployed on physical robots, to react to unplanned changes in the environment as they solve simple tasks. That is, we've used these techniques to build closed-loop systems rather than open-loop ones as before. The simulator need not match the real-world in appearance or dynamics; instead, we randomize relevant aspects of the environment, from friction to action delays to sensor noise. Our new results provide more evidence that general-purpose robots can be built by training entirely in simulation, followed by a small amount of self-calibration in the real world. This robot was trained in simulation with dynamics randomization to push a puck to a goal.

It's been a few years since I was first exposed to reinforcement learning. What got me into it was seeing this video that had trained a neural network to play Mario. As someone who grew up playing Mario, seeing deep learning being applied to something I knew so well seemed to provided the perfect introduction to the topic. Sadly though, the project was written using Torch, and I was still a naive young programmer. I didn't get too far along before the frustrations with learning lua lead me to give up, and just focus on other projects instead.

AI is transforming every industry it touches from healthcare, to retail and advertising, finance, transport, education, agriculture and so many more. To take care of all the mundane tasks employees currently handle, freeing up their time to be more creative and perform the work that machines cannot do. Today, the rapidly advancing technology is used mostly by large enterprises through machine learning and predictive analytics. AI is not a technology of the future, it's happening now, and companies who fail to adopt it will get left behind. This paper will explore the application of AI in business with research contributions from leading minds in the field including Ankur Handa, Research Scientist, OpenAI, Ian Goodfellow, Senior Research Scientist, Google Brain.

The functions of proteins and RNAs are determined by a myriad of interactions between their constituent residues, but most quantitative models of how molecular phenotype depends on genotype must approximate this by simple additive effects. While recent models have relaxed this constraint to also account for pairwise interactions, these approaches do not provide a tractable path towards modeling higher-order dependencies. Here, we show how latent variable models with nonlinear dependencies can be applied to capture beyond-pairwise constraints in biomolecules. We present a new probabilistic model for sequence families, DeepSequence, that can predict the effects of mutations across a variety of deep mutational scanning experiments significantly better than site independent or pairwise models that are based on the same evolutionary data. The model, learned in an unsupervised manner solely from sequence information, is grounded with biologically motivated priors, reveals latent organization of sequence families, and can be used to extrapolate to new parts of sequence space.

We propose generative neural network methods to generate DNA sequences and tune them to have desired properties. We present three approaches: creating synthetic DNA sequences using a generative adversarial network; a DNA-based variant of the activation maximization ("deep dream") design method; and a joint procedure which combines these two approaches together. We show that these tools capture important structures of the data and, when applied to designing probes for protein binding microarrays, allow us to generate new sequences whose properties are estimated to be superior to those found in the training data. We believe that these results open the door for applying deep generative models to advance genomics research.

Deep generative models are reported to be useful in broad applications including image generation. Repeated inference between data space and latent space in these models can denoise cluttered images and improve the quality of inferred results. However, previous studies only qualitatively evaluated image outputs in data space, and the mechanism behind the inference has not been investigated. The purpose of the current study is to numerically analyze changes in activity patterns of neurons in the latent space of a deep generative model called a "variational auto-encoder" (VAE). What kinds of inference dynamics the VAE demonstrates when noise is added to the input data are identified. The VAE embeds a dataset with clear cluster structures in the latent space and the center of each cluster of multiple correlated data points (memories) is referred as the concept. Our study demonstrated that transient dynamics of inference first approaches a concept, and then moves close to a memory. Moreover, the VAE revealed that the inference dynamics approaches a more abstract concept to the extent that the uncertainty of input data increases due to noise. It was demonstrated that by increasing the number of the latent variables, the trend of the inference dynamics to approach a concept can be enhanced, and the generalization ability of the VAE can be improved.

Researchers at OpenAI have launched a library of tools that can help researchers build faster, more efficient neural networks that take up less memory on GPUs. Neural networks are made up of layers of connected nodes. The architecture for these networks are highly variable depending on the data and application, but all models are limited by the way they run on GPUs. One way to train larger models for less computation is to introduce sparse matrices. A matrix is considered sparse if it is filled with mostly zeroes.

Elon Musk believes it's highly likely that artificial intelligence (AI) will be a threat to people. The Tesla founder is concerned that a handful of major companies will end up in control of AI systems with "extreme" levels of power. In Mr Musk's opinion, there's a very small chance that humans will be safe from such systems. "Maybe there's a five to 10 percent chance of success [of making AI safe]," he told Neuralink staff after showing them a documentary on AI, reports Rolling Stone. He also told them that he invested in DeepMind in order to keep an eye on Google's development of AI.

Human-like artificial intelligence is still a long way off, but Greg Brockman believes the time to start thinking about its safety is now. That's why, after helping to build the online-payments firm Stripe, he cofounded OpenAI along with Elon Musk and others. The nonprofit research group focuses on making sure AI continues to benefit humanity even as it increases in sophistication. Brockman plays many roles at the firm, from recruiting to helping researchers test new learning algorithms. In the long term, he says, a general AI system will need something akin to a sense of shame to prevent it from misbehaving.

Deep generative models learn a mapping from a low dimensional latent space to a high-dimensional data space. Under certain regularity conditions, these models parameterize nonlinear manifolds in the data space. In this paper, we investigate the Riemannian geometry of these generated manifolds. First, we develop efficient algorithms for computing geodesic curves, which provide an intrinsic notion of distance between points on the manifold. Second, we develop an algorithm for parallel translation of a tangent vector along a path on the manifold. We show how parallel translation can be used to generate analogies, i.e., to transport a change in one data point into a semantically similar change of another data point. Our experiments on real image data show that the manifolds learned by deep generative models, while nonlinear, are surprisingly close to zero curvature. The practical implication is that linear paths in the latent space closely approximate geodesics on the generated manifold. However, further investigation into this phenomenon is warranted, to identify if there are other architectures or datasets where curvature plays a more prominent role. We believe that exploring the Riemannian geometry of deep generative models, using the tools developed in this paper, will be an important step in understanding the high-dimensional, nonlinear spaces these models learn.