Mobile network that millions of people use every day is one of the most complex systems in real world. Optimization of mobile network to meet exploding customer demand and reduce CAPEX/OPEX poses greater challenges than in prior works. Learning to solve complex problems in real world to benefit everyone and make the world better has long been ultimate goal of AI. However, it still remains an unsolved problem for deep reinforcement learning (DRL), given imperfect information in real world, huge state/action space, lots of data needed for training, associated time/cost, multi-agent interactions, potential negative impact to real world, etc. To bridge this reality gap, we proposed a DRL framework to direct transfer optimal policy learned from multi-tasks in source domain to unseen similar tasks in target domain without any further training in both domains. First, we distilled temporal-spatial relationships between cells and mobile users to scalable 3D image-like tensor to best characterize partially observed mobile network. Second, inspired by AlphaGo, we used a novel self-play mechanism to empower DRL agent to gradually improve its intelligence by competing for best record on multiple tasks. Third, a decentralized DRL method is proposed to coordinate multi-agents to compete and cooperate as a team to maximize global reward and minimize potential negative impact. Using 7693 unseen test tasks over 160 unseen simulated mobile networks and 6 field trials over 4 commercial mobile networks in real world, we demonstrated the capability of our approach to direct transfer the learning from one simulator to another simulator, and from simulation to real world. This is the first time that a DRL agent successfully transfers its learning directly from simulation to very complex real world problems with incomplete and imperfect information, huge state/action space and multi-agent interactions.

In order to understand the formation of social conventions we need to know the specific role of control and learning in multi-agent systems. To advance in this direction, we propose, within the framework of the Distributed Adaptive Control (DAC) theory, a novel Control-based Reinforcement Learning architecture (CRL) that can account for the acquisition of social conventions in multi-agent populations that are solving a benchmark social decision-making problem. Our new CRL architecture, as a concrete realization of DAC multi-agent theory, implements a low-level sensorimotor control loop handling the agent's reactive behaviors (pre-wired reflexes), along with a layer based on model-free reinforcement learning that maximizes long-term reward. We apply CRL in a multi-agent game-theoretic task in which coordination must be achieved in order to find an optimal solution. We show that our CRL architecture is able to both find optimal solutions in discrete and continuous time and reproduce human experimental data on standard game-theoretic metrics such as efficiency in acquiring rewards, fairness in reward distribution and stability of convention formation.

Existing multi-agent reinforcement learning methods are limited typically to a small number of agents. When the agent number increases largely, the learning becomes intractable due to the curse of the dimensionality and the exponential growth of user interactions. In this paper, we present Mean Field Reinforcement Learning where the interactions within the population of agents are approximated by those between a single agent and the average effect from the overall population or neighboring agents; the interplay between the two entities is mutually reinforced: the learning of the individual agent's optimal policy depends on the dynamics of the population, while the dynamics of the population change according to the collective patterns of the individual policies. We develop practical mean field Q-learning and mean field Actor-Critic algorithms and analyze the convergence of the solution. Experiments on resource allocation, Ising model estimation, and battle game tasks verify the learning effectiveness of our mean field approaches in handling many-agent interactions in population.

Collaboration requires coordination, and we coordinate by anticipating our teammates' future actions and adapting to their plan. In some cases, our teammates' actions early on can give us a clear idea of what the remainder of their plan is, i.e. what action sequence we should expect. In others, they might leave us less confident, or even lead us to the wrong conclusion. Our goal is for robot actions to fall in the first category: we want to enable robots to select their actions in such a way that human collaborators can easily use them to correctly anticipate what will follow. While previous work has focused on finding initial plans that convey a set goal, here we focus on finding two portions of a plan such that the initial portion conveys the final one. We introduce $t$-\ACty{}: a measure that quantifies the accuracy and confidence with which human observers can predict the remaining robot plan from the overall task goal and the observed initial $t$ actions in the plan. We contribute a method for generating $t$-predictable plans: we search for a full plan that accomplishes the task, but in which the first $t$ actions make it as easy as possible to infer the remaining ones. The result is often different from the most efficient plan, in which the initial actions might leave a lot of ambiguity as to how the task will be completed. Through an online experiment and an in-person user study with physical robots, we find that our approach outperforms a traditional efficiency-based planner in objective and subjective collaboration metrics.

Learning from expert demonstrations has received a lot of attention in artificial intelligence and machine learning. The goal is to infer the underlying reward function that an agent is optimizing given a set of observations of the agent's behavior over time in a variety of circumstances, the system state trajectories, and a plant model specifying the evolution of the system state for different agent's actions. The system is often modeled as a Markov decision process, that is, the next state depends only on the current state and agent's action, and the the agent's choice of action depends only on the current state. While the former is a Markovian assumption on the evolution of system state, the later assumes that the target reward function is itself Markovian. In this work, we explore learning a class of non-Markovian reward functions, known in the formal methods literature as specifications. These specifications offer better composition, transferability, and interpretability. We then show that inferring the specification can be done efficiently without unrolling the transition system. We demonstrate on a 2-d grid world example.

This is a list of key companies worldwide with products, services, and applications in the fields related to the Artificial Intelligence (AI) & Optimization. Feel free to submit a listing and maintain it for your own business. The listing service is free. Any additional questions and comments should be submitted here. Typical AI fields include, but not limited to: Machine Learning (ML), Deep Learning, Cognitive Computing, Natural Language Processing (NLP), Computer Vision, Pattern Recognition, Autonomous Agents and Multi-Agent Systems, Automated Planning and Scheduling, Robotics, Predictive Analytics, etc.

First, let's explore the latest research on "The social and cultural roots of whale and dolphin brains" published in Nature. One of the unsolved problems of AGI research is our lack of understanding of the definition of "Generalization". I've pointed this out in my previous writing. Most of the definitions created for "Generalization" is incomplete. Definitions are either too narrow or even worse incorrect.

The University of Southampton in the U.K. has launched the Center for Machine Intelligence to develop a coherent approach to research and technology transfer. The University of Southampton in the U.K. recently launched the Center for Machine Intelligence (CMI), bringing together researchers and practitioners in artificial intelligence, machine learning, and autonomous systems to develop a coherent approach to research and technology transfer. Discussions at the launch event focused on these various technologies' application in large-scale Internet of Things systems and in the insurance and social care sectors. Research groups within the CMI will focus on the theoretical aspects of machine intelligence, including the Agents, Interaction, and Complexity group, and the Vision, Learning, and Control group. "The formation of the CMI is an important next step at a time of great advances in this field and we look forward to working with industry, policymakers and the general public as we address both national and global challenges," says Southampton professor Sarvapali Ramchurn, who will head the CMI.

IPsoft is, in many ways, an unusual entrant into the crowded, but burgeoning, artificial intelligence industry. First of all, it is not a startup, but a 20-year-old company and its leader isn't some millennial savant, but a fashionable former NYU professor named Chetan Dube. It bills its cognitive agent, Amelia, as the "world's most human AI." It got its start building and selling autonomic IT solutions and its years of experience providing business solutions give it a leg up on many of its competitors. It can offer not only technological solutions, but the insights it has gained helping businesses to streamline operations with automation.

Blockchain systems and ubiquitous computing are changing the way we do business and lead our lives. Its easy to compare blockchain with multi-agents systems communication a technology, which provides a way for multiple interacting intelligent agents (kind of swarm intelligence) to communicate with each other and with environment. Blockchain is a mystery story that provides the foundation for cryptocurrencies like Bitcoin. Some time I get confused with my school time memories data structure; ehe way blockchain is represented as a singly linked list. Each block has a hash of the previous block which can be thought of as a pointer to previous block.