We propose the Wasserstein Auto-Encoder (WAE)---a new algorithm for building a generative model of the data distribution. WAE minimizes a penalized form of the Wasserstein distance between the model distribution and the target distribution, which leads to a different regularizer than the one used by the Variational Auto-Encoder (VAE). This regularizer encourages the encoded training distribution to match the prior. We compare our algorithm with several other techniques and show that it is a generalization of adversarial auto-encoders (AAE). Our experiments show that WAE shares many of the properties of VAEs (stable training, encoder-decoder architecture, nice latent manifold structure) while generating samples of better quality, as measured by the FID score.

We study reinforcement learning (RL) in a setting with a network of agents whose states and actions interact in a local manner where the objective is to find localized policies such that the (discounted) global reward is maximized. A fundamental challenge in this setting is that the state-action space size scales exponentially in the number of agents, rendering the problem intractable for large networks. In this paper, we propose a Scalable Actor-Critic (SAC) framework that exploits the network structure and finds a localized policy that is a $O(\rho^\kappa)$-approximation of a stationary point of the objective for some $\rho\in(0,1)$, with complexity that scales with the local state-action space size of the largest $\kappa$-hop neighborhood of the network.

Traditional Reinforcement Learning (RL) algorithms either predict rewards with value functions or maximize them using policy search. We study an alternative: Upside-Down Reinforcement Learning (Upside-Down RL or UDRL), that solves RL problems primarily using supervised learning techniques. Many of its main principles are outlined in a companion report [34]. Here we present the first concrete implementation of UDRL and demonstrate its feasibility on certain episodic learning problems. Experimental results show that its performance can be surprisingly competitive with, and even exceed that of traditional baseline algorithms developed over decades of research.

-- In this paper, we propose SwarmNet - a neural network architecture that can learn to predict and imitate the behavior of an observed swarm of agents in a centralized manner . T ested on artificially generated swarm motion data, the network achieves high levels of prediction accuracy and imitation authenticity. We compare our model to previous approaches for modelling interaction systems and show how modifying components of other models gradually approaches the performance of ours. Finally, we also discuss an extension of SwarmNet that can deal with nondeterministic, noisy, and uncertain environments, as often found in robotics applications. Multi-Robot Systems (MRS) [1] describe groups of robotic agents that collectively perform complex tasks in a distributed and parallel manner through repeated interactions among each other and the environment. Such systems have attracted considerable attention in recent years with remarkable successes in a number of application domains, including defense, agriculture, logistics, disaster management, and entertainment. In particular, today's fast-paced online economy is largely fuelled by tens of thousands of warehouse robots that transport millions of items across fulfillment centers all over the world. Despite this progress, programming groups of robots to perform a joint task is still considered a complex, time-consuming, and extremely challenging endeavour. One prominent formalism for the specification of MRS is based on the identification of cost functions [2] governing the group behavior. However, this approach is not intuitive and requires a deep understanding of complex theoretical concepts across a number of mathematical fields, e.g., graph theory, manifold theory, nonlinear optimization, etc. In addition, the real-world ramifications of even small changes in a given cost function are extremely difficult to foresee.

In this paper, we propose an elegant solution that is directly addressing the bottlenecks of the traditional deep learning approaches and offers a clearly explainable internal architecture that can outperform the existing methods, requires very little computational resources (no need for GPUs) and short training times (in the order of seconds). The proposed approach, xDNN is using prototypes. Prototypes are actual training data samples (images), which are local peaks of the empirical data distribution called typicality as well as of the data density. This generative model is identified in a closed form and equates to the pdf but is derived automatically and entirely from the training data with no user- or problem-specific thresholds, parameters or intervention. The proposed xDNN offers a new deep learning architecture that combines reasoning and learning in a synergy. It is non-iterative and non-parametric, which explains its efficiency in terms of time and computational resources. From the user perspective, the proposed approach is clearly understandable to human users. We tested it on some well-known benchmark data sets such as iRoads and Caltech-256. xDNN outperforms the other methods including deep learning in terms of accuracy, time to train and offers a clearly explainable classifier. In fact, the result on the very hard Caltech-256 problem (which has 257 classes) represents a world record.

Gary Marcus thinks that symbol-manipulation is critical for causality. In biology, in a complex creature such as a human, one finds many different brain areas. Expecting a monolithic architecture to replicate that seems to Gary Marcus deeply unrealistic. Yoshua Bengio believes that sequential reasoning can be performed while staying in a deep learning framework which makes use of attention mechanisms and the injection of new modularity and training framework (e.g. Bringing causality, in something like the rich form in which it is expressed in humans, into deep learning, would be a real and lasting contribution to general artificial intelligence.

Artificial Intelligence (AI) has the ability to innovate and advance conventional practices and business operations. To bring AI to the edge, ADLINK takes a heterogeneous approach and offers a comprehensive solution portfolio of deep learning platforms and servers including acceleration engines, inference platforms, and training servers to infuse the power of AI into the smart manufacturing, smart city, logistics and warehousing, telecommunications applications and more.

Inflammation is a hallmark of many health conditions, but quantifying how the underlying biology of inflammation contributes to specific diseases has been difficult. For the first time, UNC School of Medicine researchers and colleagues now report the development of a new technology to identify white blood cells called neutrophils that are primed to eject inflammatory DNA into the circulation via a process called NETosis. The findings, published in Scientific Reports, mark the first time scientists have used machine learning tools for rapid quantitative and qualitative cell analysis in basic science. "This new test will allow investigators to measure NETosis in different diseases and to test drugs that may inhibit or promote the process," said senior author Leslie Parise, Ph.D., professor and chair of the UNC Department of Biochemistry and Biophysics. When foreign invaders such as viruses or bacteria enter our bodies, white blood cells rush in to fight the invaders in various ways.

Long considered "America's favorite pastime," MLB has been a leading sport in the adoption of advanced statistics, often called "Sabermetrics," a term coined by statistician Bill James and made famous by the book and movie "Moneyball." But after 143 years of statistic tracking through manual processes, it was time for MLB to eliminate this time-intensive practice by automating data collection and storing it in the cloud, unlocking possibilities never seen before. MLB Statcast, built by data scientists in the Amazon ML Solutions Lab, used Amazon SageMaker to train and deploy a deep neural network to predict stolen base success in real-time by using numerous data including runner's speed and burst, catcher's pop time, pitcher's velocity and handedness, lead-off distance, and the game situation; to predict the change in expected batting average as a batter steps up to the plate based on infield position shift; and to shed light on matchups with pitchers new to a batter based on deep learning pattern recognition.

This post aims to explain the concept of Word2vec and the mathematics behind the concept in an intuitive way while implementing Word2vec embedding using Gensim in Python. The basic idea of Word2vec is that instead of representing words as one-hot encoding (countvectorizer / tfidfvectorizer) in high dimensional space, we represent words in dense low dimensional space in a way that similar words get similar word vectors, so they are mapped to nearby points. Word2vec is not deep neural network, it turns text into a numeric form that deep neural network can process as input. For example, we can use "artificial" to predict "intelligence". However, the prediction itself is not our goal.