Alan Carlin

ANALYZING MYOPIC APPROACHES FOR MULTI-AGENT COMMUNICATION

Choosing when to communicate is a fundamental problem in multi-agent systems. This problem becomes particularly hard when communication is constrained and each agent has different partial information about the overall situation. Although computing the exact value of communication is intractable, it has been estimated using a standard myopic assumption. However, this assumption - that communication is only possible at the present time ntroduces error that can lead to poor agent behavior. We examine specific situations in which the myopic approach performs poorly and demonstrate an alternate approach that relaxes the assumption to improve the performance. The results provide an effective method for value-driven communication policies in multi-agent systems.

Myopic and Non-myopic Communication under Partial Observability

In decentralized settings with partial observability, agents can often benefit from communicating, but communication resources may be limited and costly. Current approaches tend to dismiss or underestimate this cost, resulting in over-communication. This paper presents a general framework to compute the value of communicating from each agent’s local perspective, by comparing the expected reward with and without communication. In order to obtain these expectations, each agent must reason about the state and belief states of the other agents, both before and after communication. We show how this can be done in the context of decentralized POMDPs and discuss ways to mitigate a common myopic assumption, where agents tend to over-communicate because they overlook the possibility that communication can be deferred or initiated by the other agents. The paper presents a theoretical framework to precisely quantify the value of communication and an effective algorithm to manage communication. Experimental results show that our approach performs well compared to other techniques suggested in the literature.

Training organizational supervisors to detect and prevent cyber insider threats: two approaches

Cyber insider threat is intentional theft from, or sabotage of, a cyber system by someone within the organization. This article explores the use of advanced cognitive and instructional principles to accelerate learning in organizational supervisors to mitigate the cyber threat. It examines the potential advantage of using serious games to engage supervisors. It also posits two systematic instructional approaches for this training challenge – optimal path modelling and a competency-based approach. The paper concludes by discussing challenges of evaluating training for seldom occurring real world phenomena, like detecting a cyber-insider threat.

Multimodal interaction for human-robot teams

Unmanned ground vehicles have the potential for supporting small dismounted teams in mapping facilities, maintaining security in cleared buildings, and extending the team’s reconnaissance and persistent surveillance capability. In order for such autonomous systems to integrate with the team, we must move beyond current interaction methods using heads-down teleoperation which require intensive human attention and affect the human operator’s ability to maintain local situational awareness and ensure their own safety. This paper focuses on the design, development and demonstration of a multimodal interaction system that incorporates naturalistic human gestures, voice commands, and a tablet interface. By providing multiple, partially redundant interaction modes, our system degrades gracefully in complex environments and enables the human operator to robustly select the most suitable interaction method given the situational demands. For instance, the human can silently use arm and hand gestures for commanding a team of robots when it is important to maintain stealth. The tablet interface provides an overhead situational map allowing waypoint-based navigation for multiple ground robots in beyond-line-of-sight conditions. Using lightweight, wearable motion sensing hardware either worn comfortably beneath the operator’s clothing or integrated within their uniform, our non-vision-based approach enables an accurate, continuous gesture recognition capability without line-of-sight constraints. To reduce the training necessary to operate the system, we designed the interactions around familiar arm and hand gestures.

Higher Automated Learning through Principal Component Analysis and Markov Models

This paper reports a hybrid method for data-driven instructional design, a method that combines Principle Components Analysis (PCA), Hidden Markov Models (HMM), and Item Response Theory (IRT). PCA is used to identify instructional objectives as well as potential student states, HMMs are used to identify dynamics between states, and IRT is used to construct measurements of state. We report on the architecture of the system along with preliminary results.

Bounded Rationality in Multiagent Systems Using Decentralized Metareasoning

Metareasoning has been used as a means for achieving bounded rationality by optimizing the tradeoff between the cost and value of the decision making process. Effective monitoring techniques have been developed to allow agents to stop their computation at the “right” time so as to optimize the overall time-dependent utility of the decision. However, these methods were designed for a single decision maker. In this chapter, we analyze the problems that arise when several agents solve components of a larger problem, each using an anytime algorithm. Metareasoning is more challenging in this case because each agent is uncertain about the progress made so far by the others. We develop a formal framework for decentralized monitoring of decision making, establish the complexity of several interesting variants of the problem, and propose solution techniques for each case.

Coordinating with Humans by Adjustable-Autonomy for Multirobot Pursuit (CHAMP)

One of the primary challenges facing the modern small-unit tactical team is the ability of the unit to safely and effectively search, explore, clear and hold urbanized terrain that includes buildings, streets, and subterranean dwellings. Buildings provide cover and concealment to an enemy and restrict the movement of forces while diminishing their ability to engage the adversary. The use of robots has significant potential to reduce the risk to tactical teams and dramatically force multiply the small units footprint. Despite advances in robotic mobility, sensing capabilities, and human-robot interaction, the use of robots in room clearing operations remains nascent. CHAMP is a software system in development that integrates with a team of robotic platforms to enable them to coordinate with a human operator performing a search and pursuit task. In this way, the human operator can either give control to the robots to search autonomously, or can retain control and direct the robots where needed. CHAMPs autonomy is built upon a combination of adversarial pursuit algorithms and dynamic function allocation strategies that maximize the teams resources. Multi-modal interaction with CHAMP is achieved using novel gesture-recognition based capabilities to reduce the need for heads-down tele-operation. The Champ Coordination Algorithm addresses dynamic and limited team sizes, generates a novel map of the area, and takes into account mission goals, user preferences and team roles. In this paper we show results from preliminary simulated experiments and find that the CHAMP system performs faster than traditional search and pursuit algorithms.

Reusing information for high-level fusion: characterizing bias and uncertainty in human-generated intelligence

To expedite the intelligence collection process, analysts reuse previously collected data. This poses the risk of analysis failure, because these data are biased in ways that the analyst may not know. Thus, these data may be incomplete, inconsistent or incorrect, have structural gaps and limitations, or simply be too old to accurately represent the current state of the world. Incorporating human-generated intelligence within the high-level fusion process enables the integration of hard (physical sensors) and soft information (human observations) to extend the ability of algorithms to associate and merge disparate pieces of information for a more holistic situational awareness picture. However, in order for high-level fusion systems to manage the uncertainty in soft information, a process needs to be developed for characterizing the sources of error and bias specific to human-generated intelligence and assessing the quality of this data. This paper outlines an approach Towards Integration of Data for unBiased Intelligence and Trust (TID-BIT) that implements a novel Hierarchical Bayesian Model for high-level situation modeling that allows the analyst to accurately reuse existing data collected for different intelligence requirements. TID-BIT constructs situational, semantic knowledge graphs that links the information extracted from unstructured sources to intelligence requirements and performs pattern matching over these attributed-network graphs for integrating information. By quantifying the reliability and credibility of human sources, TID-BIT enables the ability to estimate and account for uncertainty and bias that impact the high-level fusion process, resulting in improved situational awareness.

Decision Support in Organizations: A Case for OrgPOMDPs

In todays world, organizations are faced with increasingly large and complex problems that require decision-making under uncertainty. Current methods for optimizing such decisions fall short of handling the problem scale due to not exploiting the inherent structure of the organizations. We propose a new model called the \text it{OrgPOMDP} (Organizational POMDP), which is based on the partially observable Markov decision process (POMDP). This new model combines two powerful representations for modeling large scale problems: hierarchical modeling and factored representations. In this paper we make three key contributions: (a) Introduce the OrgPOMDP model, (b) Present an algorithm to solve OrgPOMDP problems efficiently, and (c) Apply OrgPOMDPs to scenarios in an existing large organization, the Air and Space Operation Center (AOC). We conduct experiments and show that our OrgPOMDP approach results in greater scalability and greatly reduced runtime. In fact, as the size of the problem increases, we soon reach a point at which the OrgPOMDP approach continues to provide solutions while traditional POMDP methods cannot. We also provide an empirical evaluation to highlight the benefits of an organization implementing an OrgPOMDP policy.