| 1. |
- Aler Tubella, Andrea, 1990-, et al.
(författare)
-
Interrogating the black box : Transparency through information-seeking dialogues
- 2021
-
Ingår i: Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS. - : International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS). - 9781713832621 ; , s. 106-114
-
Konferensbidrag (refereegranskat)abstract
- This paper is preoccupied with the following question: given a (possibly opaque) learning system, how can we understand whether its behaviour adheres to governance constraints? The answer can be quite simple: we just need to “ask” the system about it. We propose to construct an investigator agent to query a learning agent- the suspect agent- to investigate its adherence to a given ethical policy in the context of an information-seeking dialogue, modeled in formal argumentation settings. This formal dialogue framework is the main contribution of this paper. Through it, we break down compliance checking mechanisms into three modular components, each of which can be tailored to various needs in a vast amount of ways: an investigator agent, a suspect agent, and an acceptance protocol determining whether the responses of the suspect agent comply with the policy. This acceptance protocol presents a fundamentally different approach to aggregation: rather than using quantitative methods to deal with the non-determinism of a learning system, we leverage the use of argumentation semantics to investigate the notion of properties holding consistently. Overall, we argue that the introduced formal dialogue framework opens many avenues both in the area of compliance checking and in the analysis of properties of opaque systems.
|
|
| 2. |
|
|
| 3. |
- Bhuyan, Monowar, et al.
(författare)
-
Argumentation-based adversarial regression with multiple learners
- 2022
-
Ingår i: 2022 IEEE 34th international conference on tools with artificial intelligence (ICTAI). - : IEEE. - 9798350397444 - 9798350397451 ; , s. 96-104
-
Konferensbidrag (refereegranskat)abstract
- Despite the extensive benefits of machine learning techniques in practice, several studies demonstrated that many approaches are vulnerable to attacks. These attacks generate adversarial data to manipulate learning models that result ambiguous decisions. In this paper, we propose a hybrid-reasoning framework that combines data-driven and non-monotonic reasoning, specifically formal argumentation and adversarial regression with multiple learners, to deal with ambiguous predictions of predictive models. The introduced hybrid-reasoning framework ensures three significant benefits. It (i) provides an argumentation-based aggregation function for combining multiple learners, (ii) reduces the effort to resolve conflicts in predictions, and (iii) cost-effective and robust training in adversarial regression solutions. To illustrate the introduced framework, we consider a benchmark of resource traces obtained from Yahoo's service cluster for anomaly detection under adversarial settings. The experimental analysis shows 3% more accuracy in prediction under argumentation-based adversarial settings.
|
|
| 4. |
- Brännström, Andreas, et al.
(författare)
-
Towards Human-Aware Epistemic Planning For Promoting Behavior-Change
- 2020
-
Ingår i: Workshop Program.
-
Konferensbidrag (refereegranskat)abstract
- This paper introduces an approach to human-aware epistemicplanning in which a rational intelligent agent plans its actionsfor encouraging a human to proceed in a social virtual real-ity (VR) environment. In order to persuade the human user toexecute specific actions, the agent adapts the virtual environ-ment by adjusting motivators in the environment. The agent’smodel of the human is based on the theory of planned behav-ior (TPB), a cognitive theory to explain and predict humanbehavior. The intelligent agent manipulates the environment,a process where the agent conducts epistemic actions, i.e.,adapting the environment and observing human responses, inorder to understand the human’s behavior and encourage hu-man actions. An action reasoning framework is introducedthat defines transitions between goal-oriented human activi-ties in the virtual scenario. The proposed human-aware plan-ning architecture can also be applied in environments that arenot virtual, by utilizing modern mobile devices which havebuilt-in sensors that measure motion, orientation, and variousenvironmental conditions.
|
|
| 5. |
|
|
| 6. |
- Ingeson, Martin, et al.
(författare)
-
Microsoft Hololens : A mHealth Solution for Medication Adherence
- 2019
-
Ingår i: Artificial Intelligence in Health. - Cham : Springer. - 9783030127374 - 9783030127381 ; , s. 99-115
-
Konferensbidrag (refereegranskat)abstract
- The aim of this paper is to introduce a smart mHealth application based on the augmented reality (AR)-paradigm that can support patients with common problems, related to management of their medication. This smart mHealth application is designed and implemented as a medication coach intelligent agent, called Medication Coach Intelligent Agent (MCIA). The MCIA has to manage different types of information such as the medication plan (medication regime) of the patients, medication restrictions, as well as the patient’s preferences and sensor input data from an AR-headset. Considering all this information, the MCIA leads with holistic decisions in order to offer personalized and unobtrusive interventions, in an autonomous way, to the patients. From a long-term perspective, the MCIA should also evaluate its performance over time and adapt in order to improve its interventions with the patients. To show the feasibility of our approach, a proof-of-concept prototype was implemented and evaluated. In this proof-of-concept prototype, the MCIA has been embodied as a smart augmented reality (AR)-mHealth application in the settings of a Microsoft HoloLens. The results show a high potential for using the MCIA in real settings.
|
|
| 7. |
- Ingeson, Martin, et al.
(författare)
-
Smart augmented reality mHealth for medication adherence
- 2018
-
Ingår i: AIH 2018.
-
Konferensbidrag (refereegranskat)abstract
- The aim of this paper is to introduce a smart mHealth applicationbased on the augmented reality (AR)-paradigm that can support patients with common problems, related to management of their medication. This smart mHealth application is designed and implemented as a medication coach intelligent agent, called Medication Coach Intelligent Agent (MCIA). TheMCIA will have to manage different types of information such as the medication plan (medication regime) of the patients, medication restrictions, as well as the patient’s preferences andsensor input data from an AR-headset. Considering all this information, theMCIA leads with holistic decisions in order to offer personalized and unobtrusive inter-ventions, in an autonomous way, to the patients. From a long term perspective, the MCIAshould also evaluate its performance over time and adapt in order to improve its interventions with the patients. To show the feasibility of our approach,a proof-of-concept prototype was implemented and evaluated. The results showa high potential for using theMCIAin real settings.
|
|
| 8. |
- Kampik, Timotheus, 1989-, et al.
(författare)
-
Toward consistent agreement approximation in abstract argumentation and beyond
- 2021
-
Ingår i: Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS. - : International Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS). - 9781450383073 ; , s. 1551-1553
-
Konferensbidrag (refereegranskat)abstract
- In cooperative human decision-making, agreements are often not total; a partial degree of agreement is sufficient to commit to a decision and move on, as long as one is somewhat confident that the involved parties are likely to stand by their commitment in the future, given no drastic unexpected changes. In this work, we introduce models that allow autonomous agents to reach such agreements, using abstract argumentation as the underlying model.
|
|
| 9. |
- Lindgren, Helena, et al.
(författare)
-
Computer-Supported Assessment for Tailoring Assistive Technology
- 2016
-
Ingår i: DH'16: PROCEEDINGS OF THE 2016 DIGITAL HEALTH CONFERENCE. - New York : Association for Computing Machinery (ACM). - 9781450342247 ; , s. 1-10
-
Konferensbidrag (refereegranskat)abstract
- The main purpose of assistive technology is to support an individual's daily activities, in order to increase ability, autonomy, relatedness and quality of life. The aim for the work presented in this article is to develop automated methods to tailor the behavior of the assistive technology for the purpose to provide just-in-time, adaptive interventions targeting multiple domains. This requires methods for representing and updating the user model, including goals, preferences, abilities, activity and its situation. We focus the assessment and intervention tasks typically performed by therapists and provide knowledge-based technology for supporting the process. A formative evaluation study was conducted as a part of a participatory action research process, involving two rehabilitation experts, two young individuals and one senior individual as end-user participants, in addition to knowledge engineers. The main contribution of this work is a theory-based method for assessing the individual's goals, preferences, abilities and motives, which is used for building a holistic user model. The user model is continuously updated and functions as the base for tailoring the system's assistive behavior during intervention and follow-up.
|
|
| 10. |
- Mariela, Morveli-Espinoza, et al.
(författare)
-
An Argumentation-based Approach for Identifying and Dealing with Incompatibilities among Procedural Goals
- 2019
-
Ingår i: International Journal of Approximate Reasoning. - : Elsevier. - 0888-613X .- 1873-4731. ; 105, s. 1-26
-
Tidskriftsartikel (refereegranskat)abstract
- During the first step of practical reasoning, i.e. deliberation, an intelligent agent generates a set of pursuable goals and then selects which of them he commits to achieve. An intelligent agent may in general generate multiple pursuable goals, which may be incompatible among them. In this paper, we focus on the definition, identification and resolution of these incompatibilities. The suggested approach considers the three forms of incompatibility introduced by Castelfranchi and Paglieri, namely the terminal incompatibility, the instrumental or resources incompatibility and the superfluity. We characterize computationally these forms of incompatibility by means of arguments that represent the plans that allow an agent to achieve his goals. Thus, the incompatibility among goals is defined based on the conflicts among their plans, which are represented by means of attacks in an argumentation framework. We also work on the problem of goals selection; we propose to use abstract argumentation theory to deal with this problem, i.e. by applying argumentation semantics. We use a modified version of the “cleaner world” scenario in order to illustrate the performance of our proposal.
|
|
