This paper presents a micro-traffic simulation (named “DeepTraffic”) where the perception, control, and planning systems for one of the cars are all handled by a single neural network as part of a model-free, off-policy reinforcement learning process. The paper also […]
This paper discusses the best approaches to take with Genetic algorithms (GA), a branch of evolutionary computing (EC) that mimics the theory of evolution and natural selection, where the technique is based on an heuristic random search. Crossover and mutation […]
This paper addresses the important problem of discerning hateful content in social media. It proposes a detection scheme that is an ensemble of Recurrent Neural Network (RNN) classifiers, and it incorporates various features associated with user-related information, such as the […]
This paper presents a study in Distributed Deep Reinforcement Learning (DDRL) focused on scalability of a state-of-the-art Deep Reinforcement Learning algorithm known as Batch Asynchronous Advantage ActorCritic (BA3C). Using synchronous training on the node level (while keeping the local, single […]
As examples like Wikipedia, Google, and Linux illustrate, new information and communication technologies are now enabling dramatically new ways of connecting large numbers of people and computers to produce intelligent behavior. These new forms of “collective intelligence” are already having […]
Much work remains in the area of social conversation as well as free-form conversation over a broad range of domains and topics. To advance the state of the art in conversational AI, Amazon launched the Alexa Prize, a 2.5-million-dollar university […]
The subjectivity associated with evaluating conversations is a key element underlying the challenge of building non-goal oriented dialogue systems. This paper proposes a comprehensive evaluation strategy with multiple metrics designed to reduce subjectivity by selecting metrics which correlate well with […]
