How did our agents perform as well as they did? First, we noticed that the agents had very fast reaction times and were very accurate taggers, which might explain their performance (tagging is a tactical action that sends opponents back to their starting point). Humans are comparatively slow to process and act on sensory input, due to our slower biological signalling. Here's an example of a reaction time test you can try yourself. Thus, our agents' superior performance might be a result of their faster visual processing and motor control.

While there are many examples in which robots provide social assistance, a lack of theory on how the robots should decide how to assist impedes progress in realizing these technologies. To address this deficiency, we propose a pair of computational models to guide a robot as it provides social assistance. The model of social autonomy helps a robot select an appropriate assistance that will help with the task at hand while also maintaining the autonomy of the person being assisted. The model of social alliance describes how a to determine whether the robot and the person being assisted are cooperatively working towards the same goal. Each of these models are rooted in social reasoning between people, and we describe here our ongoing work to adapt this social reasoning to human-robot interactions. Socially assistive robots (SARs) provide social assistance instead of physically intervening.

The work presented in this paper aims to explore how, and to what extent, an adaptive robotic coach has the potential to provide extra motivation to adhere to long-term rehabilitation and help fill the coaching gap which occurs during repetitive solo practice in high performance sport. Adapting the behavior of a social robot to a specific user, using reinforcement learning (RL), could be a way of increasing adherence to an exercise routine in both domains. The requirements gathering phase is underway and is presented in this paper along with the rationale of using RL in this context.

Human-supervision in multi-agent teams is a critical requirement to ensure that the decision-maker's risk preferences are utilized to assign tasks to robots. In stressful complex missions that pose risk to human health and life, such as humanitarian-assistance and disaster-relief missions, human mistakes or delays in tasking robots can adversely affect the mission. To assist human decision making in such missions, we present an alert-generation framework capable of detecting various modes of potential failure or performance degradation. We demonstrate that our framework, based on state machine simulation and formal methods, offers probabilistic modeling to estimate the likelihood of unfavorable events. We introduce smart simulation that offers a computationally-efficient way of detecting low-probability situations compared to standard Monte-Carlo simulations. Moreover, for certain class of problems, our inference-based method can provide guarantees on correctly detecting task failures.

Smart speakers and robots become ever more prevalent in our daily lives. These agents are able to execute a wide range of tasks and actions and, therefore, need systems to control their execution. Current state-of-the-art such as (deep) reinforcement learning, however, requires vast amounts of data for training which is often hard to come by when interacting with humans. To overcome this issue, most systems still rely on Finite State Machines. We introduce Petri Net Machines which present a formal definition for state machines based on Petri Nets that are able to execute concurrent actions reliably, execute and interleave several plans at the same time, and provide an easy to use modelling language. We show their workings based on the example of Human-Robot Interaction in a shopping mall.

Recent years have seen a rising interest in developing AI algorithms for real world big data domains ranging from autonomous cars to personalized assistants. At the core of these algorithms are architectures that combine deep neural networks, for approximating the underlying multidimensional state-spaces, with reinforcement learning, for controlling agents that learn to operate in said state-spaces towards achieving a given objective. The talk will first outline notable past and future efforts in deep reinforcement learning as well as identify fundamental problems that this technology has been struggling to overcome. Towards mitigating these problems (and open up an alternative path to general artificial intelligence), I will then summarize a brain computing model of intelligence, rooted in the latest findings in neuroscience. The talk will conclude with an overview of the recent research efforts in the field of multi-agent systems, to provide the future teams of humans and agents with the necessary tools that allow them to safely co-exist.

Marianne and Marcus Wallenberg Foundation has granted SEK 96 million to be shared by 16 research projects studying the impact of artificial intelligence and autonomous systems on our society and our behaviour. The 16 projects seek to answer a number of questions relating to ethics, society and behaviour in the technology shift that society is facing. Examples of these questions include: How does the labour market change when robots take over certain jobs? What does the growing use of facial and voice recognition technology entail? How is human behaviour affected by the increasing use of drones?

There is a growing interest in large-scale machine learning and optimization over decentralized networks, e.g. in the context of multi-agent learning and federated learning. Due to the imminent need to alleviate the communication burden, the investigation of communication-efficient distributed optimization algorithms --- particularly for empirical risk minimization --- has flourished in recent years. A large faction of these algorithms have been developed for the master/slave setting, relying on the presence of a central parameter server that can communicate with all agents. This paper focuses on distributed optimization over the network-distributed or the decentralized setting, where each agent is only allowed to aggregate information from its neighbors over a network (namely, no centralized coordination is present). By properly adjusting the global gradient estimate via a tracking term, we develop a communication-efficient approximate Newton-type method, called Network-DANE, which generalizes DANE [Shamir et al., 2014] for decentralized networks. We establish linear convergence of Network-DANE for quadratic losses, which shed light on the impact of data homogeneity and network connectivity upon the rate of convergence. Our key algorithmic ideas can be applied, in a systematic manner, to obtain decentralized versions of other master/slave distributed algorithms. A notable example is our development of Network-SVRG, which employs stochastic variance reduction [Johnson and Zhang, 2013] at each agent to accelerate local computation. The proposed algorithms are built upon the primal formulation without resorting to the dual. Numerical evidence is provided to demonstrate the appealing performance of our algorithms over competitive baselines, in terms of both communication and computation efficiency.

This paper introduces a new negotiating agent model for automated negotiation. We focus on applications without time pressure with multidi-mensional negotiation on both continuous and discrete domains. The agent bidding strategy relies on Monte Carlo Tree Search, which is a trendy method since it has been used with success on games with high branching factor such as Go. It also exploits opponent modeling techniques thanks to Gaussian process regression and Bayesian learning. Evaluation is done by confronting the existing agents that are able to negotiate in such context: Random Walker, Tit-for-tat and Nice Tit-for-Tat. None of those agents succeeds in beating our agent. Also, the modular and adaptive nature of our approach is a huge advantage when it comes to optimize it in specific applicative contexts.

We describe the concept of logical scaffolds, which can be used to improve the quality of software that relies on AI components. We explain how some of the existing ideas on runtime monitors for perception systems can be seen as a specific instance of logical scaffolds. Furthermore, we describe how logical scaffolds may be useful for improving AI programs beyond perception systems, to include general prediction systems and agent behavior models. Keywords: AI · Autonomous systems · Formal methods. 1 Introduction Recent progress in AI has led to possible deployment in a wide variety of important domains. This includes safety-critical cyberphysical systems such as automobiles [1] and airplanes [7], but also decision making systems in diverse domains including legal [15] and military applications [3].