Since the Theano team announced that it would cease development and maintenance of Theano within a year, we, the PyMC developers, have been actively discussing what to do about this. We are very excited to announce that the new version of PyMC will use TensorFlow Probability (TFP) as its backend. TensorFlow already has a very broad user base and with TFP gained a powerful new library with elegant support for probability distributions and transformations (called bijections, see the TFP paper for a full description), as well as a layer for constructing probabilistic models, called Edward2. It is clear that TFP's focus is to provide a strong foundation upon which flexible statistical models for inference and prediction can be constructed from the ground up. Its focus is not, however, to provide a high-level API which makes construction and fitting of common classes of models easy for applied users.

Which authors of this paper are endorsers? Disable MathJax (What is MathJax?)

Deep Learning is the buzzword of the day in IT. Fueled by the significant advancements generated by GPUs and lately by FPGAs, deep learning is on the path of becoming ubiquitous. Yet most people are unaware of the fact that the first incarnation of a neural net, the perceptron, has its 60th birthday this year. Once almost deemed as a "dead end", neural nets, represented by their most preeminent incarnation – the deep learning nets, are coming back into the public spotlight with a vengeance. Join me in this session to discover the inner workings of deep learning networks, their advantages and pitfalls, as well as their areas of applicability. I'll cover the history and evolution of the field as well as its present state of the art.

Artificial intelligence (AI) technologies are quickly transforming almost every sphere of our lives. From how we communicate to the means we use for transportation, we seem to be getting increasingly addicted to them. Because of these rapid advancements, massive amounts of talent and resources are dedicated to accelerating the growth of the technologies. Here is a list of 8 best open source AI technologies you can use to take your machine learning projects to the next level. Initially released in 2015, TensorFlow is an open source machine learning framework that is easy to use and deploy across a variety of platforms. It is one of the most well-maintained and extensively used frameworks for machine learning.

Session title is "Deep Learning Lyrics" and is about machine generating lyrics based on existing lyrics with Keras and Tensorflow. Last parameter is the diversity (temperature). Higher value means more errors but maybe more "innovative" results. See also the list of contributors who participated in this project. This project is licensed under the MIT License - see the LICENSE.md

Every day, more than 1 million people in the United States require physical assistance to get dressed, whether because of injury, permanent disability, age, or other debilitating factors. To alleviate the problem, researchers from Georgia Tech built a deep learning-equipped robot that can help people get dressed. "What the robot is trying to do is to take the person's perspective of what a person is feeling during assistance," said Zachary Erickson, a robotics Ph.D. Student at Georgia Tech. "When the robot is doing this, it's using what it feels on its fingertips or its gripper and saying, what do I think a person is feeling while being dressed?" The robot, named PR2, was trained using NVIDIA Tesla V100 GPUs on the Amazon Web Services cloud with the cuDNN-accelerated Keras and TensorFlow deep learning frameworks.

Machine learning is enabling computers to tackle tasks that have, until now, only been carried out by people. The next wave of IT innovation will be powered by artificial intelligence and machine learning. We look at the ways companies can take advantage of it and how to get started. From driving cars to translating speech, machine learning is driving an explosion in the capabilities of artificial intelligence -- helping software make sense of the messy and unpredictable real world. But what exactly is machine learning and what is making the current boom in machine learning possible? At a very high level, machine learning is the process of teaching a computer system how to make accurate predictions when fed data.

We apply neural nets with ReLU gates in online reinforcement learning. Our goal is to train these networks in an incremental manner, without the computationally expensive experience replay. By studying how individual neural nodes behave in online training, we recognize that the global nature of ReLU gates can cause undesirable learning interference in each node's learning behavior. We propose reducing such interferences with two efficient input transformation methods that are geometric in nature and match well the geometric property of ReLU gates. The first one is tile coding, a classic binary encoding scheme originally designed for local generalization based on the topological structure of the input space. The second one (EmECS) is a new method we introduce; it is based on geometric properties of convex sets and topological embedding of the input space into the boundary of a convex set. We discuss the behavior of the network when it operates on the transformed inputs. We also compare it experimentally with some neural nets that do not use the same input transformations, and with the classic algorithm of tile coding plus a linear function approximator, and on several online reinforcement learning tasks, we show that the neural net with tile coding or EmECS can achieve not only faster learning but also more accurate approximations. Our results strongly suggest that geometric input transformation of this type can be effective for interference reduction and takes us a step closer to fully incremental reinforcement learning with neural nets.

We propose Trusted Neural Network (TNN) models, which are deep neural network models that satisfy safety constraints critical to the application domain. We investigate different mechanisms for incorporating rule-based knowledge in the form of first-order logic constraints into a TNN model, where rules that encode safety are accompanied by weights indicating their relative importance. This framework allows the TNN model to learn from knowledge available in form of data as well as logical rules. We propose multiple approaches for solving this problem: (a) a multi-headed model structure that allows trade-off between satisfying logical constraints and fitting training data in a unified training framework, and (b) creating a constrained optimization problem and solving it in dual formulation by posing a new constrained loss function and using a proximal gradient descent algorithm. We demonstrate the efficacy of our TNN framework through experiments using the open-source TORCS~\cite{BernhardCAA15} 3D simulator for self-driving cars. Experiments using our first approach of a multi-headed TNN model, on a dataset generated by a customized version of TORCS, show that (1) adding safety constraints to a neural network model results in increased performance and safety, and (2) the improvement increases with increasing importance of the safety constraints. Experiments were also performed using the second approach of proximal algorithm for constrained optimization --- they demonstrate how the proposed method ensures that (1) the overall TNN model satisfies the constraints even when the training data violates some of the constraints, and (2) the proximal gradient descent algorithm on the constrained objective converges faster than the unconstrained version.

In this paper, we aim at introducing a new machine learning model, namely reconciled polynomial machine, which can provide a unified representation of existing shallow and deep machine learning models. Reconciled polynomial machine predicts the output by computing the inner product of the feature kernel function and variable reconciling function. Analysis of several concrete models, including Linear Models, FM, MVM, Perceptron, MLP and Deep Neural Networks, will be provided in this paper, which can all be reduced to the reconciled polynomial machine representations. Detailed analysis of the learning error by these models will also be illustrated in this paper based on their reduced representations from the function approximation perspective.