Robotic process automation (RPA) has emerged as the leading approach to automate tasks in business processes. Moving away from back-end automation, RPA automated the mouse-click on user interfaces; this outside-in approach reduced the overhead of updating legacy software. However, its many shortcomings, namely its lack of accessibility to business users, have prevented its widespread adoption in highly regulated industries. In this work, we explore interactive automation in the form of a conversational digital assistant. It allows business users to interact with and customize their automation solutions through natural language. The framework, which creates such assistants, relies on a multi-agent orchestration model and conversational wrappers for autonomous agents including RPAs. We demonstrate the effectiveness of our proposed approach on a loan approval business process and a travel preapproval business process.

We consider the problem of dynamically allocating tasks to multiple agents under time window constraints and task completion uncertainty. Our objective is to minimize the number of unsuccessful tasks at the end of the operation horizon. We present a multi-robot allocation algorithm that decouples the key computational challenges of sequential decision-making under uncertainty and multi-agent coordination and addresses them in a hierarchical manner. The lower layer computes policies for individual agents using dynamic programming with tree search, and the upper layer resolves conflicts in individual plans to obtain a valid multi-agent allocation. Our algorithm, Stochastic Conflict-Based Allocation (SCoBA), is optimal in expectation and complete under some reasonable assumptions. In practice, SCoBA is computationally efficient enough to interleave planning and execution online. On the metric of successful task completion, SCoBA consistently outperforms a number of baseline methods and shows strong competitive performance against an oracle with complete lookahead. It also scales well with the number of tasks and agents. We validate our results over a wide range of simulations on two distinct domains: multi-arm conveyor belt pick-and-place and multi-drone delivery dispatch in a city.

In recent years, many sectors have experienced significant progress in automation, associated with the growing advances in artificial intelligence and machine learning. There are already automated robotic weapons, which are able to evaluate and engage with targets on their own, and there are already autonomous vehicles that do not need a human driver. It is argued that the use of increasingly autonomous systems (AS) should be guided by the policy of human control, according to which humans should execute a certain significant level of judgment over AS. While in the military sector there is a fear that AS could mean that humans lose control over life and death decisions, in the transportation domain, on the contrary, there is a strongly held view that autonomy could bring significant operational benefits by removing the need for a human driver. This article explores the notion of human control in the United States in the two domains of defense and transportation.

One of the ultimate goals of artificial intelligence is the ability for machines to operate on their own, with little or any human interaction. This idea of autonomous systems makes up one of the seven patterns of AI that represents the common ways that organizations are applying AI. While some of the patterns are focused on predictive analytics or conversational patterns, or systems that can recognize things in the world around us, those patterns still involve human interaction. After all, we need humans to be involved in conversational or recognition systems. However, the autonomous pattern is much more complicated as we're asking a machine to do something in the real world without a human in the loop.

The design and implementation of an e-voting system is a challenging task. Formal analysis can be of great help here. In particular, it can lead to a better understanding of how the voting system works, and what requirements on the system are relevant. In this paper, we propose that the state-of-art model checker Uppaal provides a good environment for modelling and preliminary verification of voting protocols. To illustrate this, we present an Uppaal model of Pr\^et \`a Voter, together with some natural extensions. We also show how to verify a variant of receipt-freeness, despite the severe limitations of the property specification language in the model checker.

"Nonstationarity" is a fundamental problem in cooperative multi-agent reinforcement learning (MARL)--each agent must relearn information about the other agent's policies due to the other agents learning, causing information to "ring" between agents and convergence to be slow. The MAILP model, introduced by Terry and Grammel (2020), is a novel model of information transfer during multi-agent learning. We use the MAILP model to show that increasing training centralization arbitrarily mitigates the slowing of convergence due to nonstationarity. The most centralized case of learning is parameter sharing, an uncommonly used MARL method, specific to environments with homogeneous agents, that bootstraps a single-agent reinforcement learning (RL) methods and learns an identical policy for each agent. We experimentally replicate the result of increased learning centralization leading to better performance on the MARL benchmark set from Gupta et al. (2017). We further apply parameter sharing to 8 "more modern" single-agent deep RL (DRL) methods for the first time in the literature. With this, we achieved the best documented performance on a set of MARL benchmarks and achieved up to 44 times more average reward in as little as 16% as many episodes compared to documented parameter sharing arrangement. We finally offer a formal proof of a set of methods that allow parameter sharing to serve in environments with heterogeneous agents.

Human collaborators can effectively communicate with their partners to finish a common task by inferring each other's mental states (e.g., goals, beliefs, and desires). Such mind-aware communication minimizes the discrepancy among collaborators' mental states, and is crucial to the success in human ad-hoc teaming. We believe that robots collaborating with human users should demonstrate similar pedagogic behavior. Thus, in this paper, we propose a novel explainable AI (XAI) framework for achieving human-like communication in human-robot collaborations, where the robot builds a hierarchical mind model of the human user and generates explanations of its own mind as a form of communications based on its online Bayesian inference of the user's mental state. To evaluate our framework, we conduct a user study on a real-time human-robot cooking task. Experimental results show that the generated explanations of our approach significantly improves the collaboration performance and user perception of the robot. Code and video demos are available on our project website: https://xfgao.github.io/xCookingWeb/.

In the opening pages of Burn-In , an FBI agent conducts close-quarters surveillance of a suspected terrorist bomber in Washington, D.C. Simultaneously, in New Jersey, an elderly gentleman listens attentively to the enthusiastic technological prognostications of a world-famous computer scientist and mathematician from the back of a hallowed lecture hall at Princeton University. Moments later, he bludgeons the speaker to death with his cane. In this, their second novel, coauthors Peter Warren Singer and August Cole—both renowned technology and policy experts—come close to perfecting the genre of educational and informative techno-thriller. Like their first such collaboration ([ 1 ][1]), this latest entry portrays a world in which conventional aspects of domestic security and law enforcement—combating terrorism, managing protests and social upheavals, tracking a serial killer, providing a secure environment on college campuses—all occur within a transformative technological context that both enables and simultaneously disrupts these myriad objectives. As the narrative unfolds, a complex tapestry of emergent, disruptive technologies is revealed. Far from the fanciful inventions that typically populate science fiction, the systems described herein are currently available or under development for imminent deployment. The D.C. traffic congestion with which agent Lara Keegan and her partner have to contend, for example, is mostly composed of driverless vehicles, their complex operational algorithms engaged in competitive maneuvering for even the slightest comparative advantage. If the agents invoke the emergency override protocol granted to law enforcement personnel and cause the other vehicles to move aside, the surveillance drones buzzing overhead will immediately transmit this activity to the news outlets that operate them, alerting the terrorist to their presence. Keegan's field of vision, meanwhile, is networked into an operations command center via virtual reality glasses, which display real-time data on the suspect's location. These “viz glasses” continuously exchange data with other law enforcement personnel, while simultaneously performing facial scans of the surrounding crowds, subjecting each passerby to massive digital analysis. Once apprehended, despite his uncooperative silence, the suspect's identity is unmasked by a Tactical Autonomous Mobility System (TAMS), a military robot whose combat utility proved minimal and is now being tested for possible use in domestic law enforcement scenarios. Keegan, we learn, has been selected to field-test this robotic deep-learning technology system because of her prior experience managing the deployment and “force mix” of unmanned systems for the Marine Corps in Afghanistan. In technology circles, what she has been asked to undertake is known as a burn-in, a lengthy trial run of any new technological breakthrough, designed to push it to its limits of reliable functionality. The novel also contains ample instances of what the Defense Advanced Research Projects Agency (DARPA) and the National Science Foundation dub the ethical, legal, and social implications (ELSI) of technological development and diffusion. Just before his death, for example, the Princeton computer scientist boasts to his elderly guest how his use of Linux open-source software to develop complex machine-learning algorithms has made artificial intelligence (AI) universally available and affordable for every conceivable purpose. As his killer peels off an AI-designed silicon facial mask (manufactured on a 3D printer to confuse the university's AI-assisted security and surveillance system), he reveals himself to be a former DARPA engineer whose wife and son were tragically killed in a Metro crash caused by dangerous emergent behaviors in one of the scientist's AI-governed public transportation systems. This narrative thread, and many others throughout the book, illustrate what coauthor Peter Warren Singer identified in his widely acclaimed book Wired for War (published in 2009) as a key constituent of technological innovation and advance: “Anything that can go wrong, will—at the worst possible moment.” The aim of this work of fiction is not merely to engage and entertain but also to educate and inform readers about the vast array of automated and increasingly intelligent autonomous systems that are proliferating in availability and use. The authors provide detailed documentation of the actual features and current use of these systems, together with a companion educational guide to help instructors use the novel to teach about the profound depths of the robotic and AI revolution that is taking place all around us. 1. [↵][2]1. P. W. Singer, 2. A. Cole , Ghost Fleet: A Novel of the Next World War (Houghton Mifflin Harcourt, 2015). [1]: #ref-1 [2]: #xref-ref-1-1 "View reference 1 in text"

Given the complexity of real-world, unstructured domains, it is often impossible or impractical to design models that include every feature needed to handle all possible scenarios that an autonomous system may encounter. For an autonomous system to be reliable in such domains, it should have the ability to improve its competence online. In this paper, we propose a method for improving the competence of a system over the course of its deployment. We specifically focus on a class of semi-autonomous systems known as competence-aware systems that model their own competence -- the optimal extent of autonomy to use in any given situation -- and learn this competence over time from feedback received through interactions with a human authority. Our method exploits such feedback to identify important state features missing from the system's initial model, and incorporates them into its state representation. The result is an agent that better predicts human involvement, leading to improvements in its competence and reliability, and as a result, its overall performance.

The actions of intelligent agents, such as chatbots, recommender systems, and virtual assistants are typically not fully transparent to the user. Consequently, using such an agent involves the user exposing themselves to the risk that the agent may act in a way opposed to the user's goals. It is often argued that people use trust as a cognitive shortcut to reduce the complexity of such interactions. Here we formalise this by using the methods of evolutionary game theory to study the viability of trust-based strategies in repeated games. These are reciprocal strategies that cooperate as long as the other player is observed to be cooperating. Unlike classic reciprocal strategies, once mutual cooperation has been observed for a threshold number of rounds they stop checking their co-player's behaviour every round, and instead only check with some probability. By doing so, they reduce the opportunity cost of verifying whether the action of their co-player was actually cooperative. We demonstrate that these trust-based strategies can outcompete strategies that are always conditional, such as Tit-for-Tat, when the opportunity cost is non-negligible. We argue that this cost is likely to be greater when the interaction is between people and intelligent agents, because of the reduced transparency of the agent. Consequently, we expect people to use trust-based strategies more frequently in interactions with intelligent agents. Our results provide new, important insights into the design of mechanisms for facilitating interactions between humans and intelligent agents, where trust is an essential factor.