Experiments with Multiple BDI Agents with Dynamic Learning Capabilities

Author(s):  
Amelia Bădică ◽  
Costin Bădică ◽  
Maria Ganzha ◽  
Mirjana Ivanović ◽  
Marcin Paprzycki
Author(s):  
Stéphane Airiau ◽  
Lin Padgham ◽  
Sebastian Sardina ◽  
Sandip Sen

Belief, Desire, and Intentions (BDI) agents are well suited for complex applications with (soft) real-time reasoning and control requirements. BDI agents are adaptive in the sense that they can quickly reason and react to asynchronous events and act accordingly. However, BDI agents lack learning capabilities to modify their behavior when failures occur frequently. We discuss the use of past experience to improve the agent’s behavior. More precisely, we use past experience to improve the context conditions of the plans contained in the plan library, initially set by a BDI programmer. First, we consider a deterministic and fully observable environment and we discuss how to modify the BDI agent to prevent re-occurrence of failures, which is not a trivial task. Then, we discuss how we can use decision trees to improve the agent’s behavior in a non-deterministic environment.


2016 ◽  
Vol 69 (10) ◽  
pp. 4287-4303 ◽  
Author(s):  
Martie-Louise Verreynne ◽  
Damian Hine ◽  
Len Coote ◽  
Rachel Parker

ASHA Leader ◽  
2009 ◽  
Vol 14 (5) ◽  
pp. 2-2
Author(s):  
Larry Boles ◽  
Amy J. Hadley ◽  
Jeanne M. Johnson ◽  
Joan A. Luckhurst ◽  
Christine Krkovich

2020 ◽  
Author(s):  
Amy K. Clark ◽  
Meagan Karvonen

Alternate assessments based on alternate achievement standards (AA-AAS) have historically lacked broad validity evidence and an overall evaluation of the extent to which evidence supports intended uses of results. An expanding body of validation literature, the funding of two AA-AAS consortia, and advances in computer-based assessment have supported improvements in AA-AAS validation. This paper describes the validation approach used with the Dynamic Learning Maps® alternate assessment system, including development of the theory of action, claims, and interpretive argument; examples of evidence collected; and evaluation of the evidence in light of the maturity of the assessment system. We focus especially on claims and sources of evidence unique to AA-AAS and especially the Dynamic Learning Maps system design. We synthesize the evidence to evaluate the degree to which it supports the intended uses of assessment results for the targeted population. Considerations are presented for subsequent data collection efforts.


Author(s):  
Yoko E. Fukumura ◽  
Julie McLaughlin Gray ◽  
Gale M. Lucas ◽  
Burcin Becerik-Gerber ◽  
Shawn C. Roll

Workplace environments have a significant impact on worker performance, health, and well-being. With machine learning capabilities, artificial intelligence (AI) can be developed to automate individualized adjustments to work environments (e.g., lighting, temperature) and to facilitate healthier worker behaviors (e.g., posture). Worker perspectives on incorporating AI into office workspaces are largely unexplored. Thus, the purpose of this study was to explore office workers’ views on including AI in their office workspace. Six focus group interviews with a total of 45 participants were conducted. Interview questions were designed to generate discussion on benefits, challenges, and pragmatic considerations for incorporating AI into office settings. Sessions were audio-recorded, transcribed, and analyzed using an iterative approach. Two primary constructs emerged. First, participants shared perspectives related to preferences and concerns regarding communication and interactions with the technology. Second, numerous conversations highlighted the dualistic nature of a system that collects large amounts of data; that is, the potential benefits for behavior change to improve health and the pitfalls of trust and privacy. Across both constructs, there was an overarching discussion related to the intersections of AI with the complexity of work performance. Numerous thoughts were shared relative to future AI solutions that could enhance the office workplace. This study’s findings indicate that the acceptability of AI in the workplace is complex and dependent upon the benefits outweighing the potential detriments. Office worker needs are complex and diverse, and AI systems should aim to accommodate individual needs.


2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Charlotte Canteloup ◽  
Mabia B. Cera ◽  
Brendan J. Barrett ◽  
Erica van de Waal

AbstractSocial learning—learning from others—is the basis for behavioural traditions. Different social learning strategies (SLS), where individuals biasedly learn behaviours based on their content or who demonstrates them, may increase an individual’s fitness and generate behavioural traditions. While SLS have been mostly studied in isolation, their interaction and the interplay between individual and social learning is less understood. We performed a field-based open diffusion experiment in a wild primate. We provided two groups of vervet monkeys with a novel food, unshelled peanuts, and documented how three different peanut opening techniques spread within the groups. We analysed data using hierarchical Bayesian dynamic learning models that explore the integration of multiple SLS with individual learning. We (1) report evidence of social learning compared to strictly individual learning, (2) show that vervets preferentially socially learn the technique that yields the highest observed payoff and (3) also bias attention toward individuals of higher rank. This shows that behavioural preferences can arise when individuals integrate social information about the efficiency of a behaviour alongside cues related to the rank of a demonstrator. When these preferences converge to the same behaviour in a group, they may result in stable behavioural traditions.


Sign in / Sign up

Export Citation Format

Share Document