Generating justifications for norm-related agent decisions

Kasenberg, Daniel; Roque, Antonio; Thielstrom, Ravenna; Chita-Tegmark, Meia; Scheutz, Matthias

doi:10.18653/v1/w19-8660

Cited by 14 publications

(6 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is an extension of the DIARC natural language understanding system (Cantrell et al, 2010). The natural language generation (NLG) process, and in particular how our system generates fairly natural-sounding utterances from temporal logic statements, is described in Kasenberg et al, 2019. This paper extends that work by (1) adding the NLU capabilities; and (2) adding norm addition/removal and "suppose" hypotheticals about the agent's norms.…”

Section: Contributionmentioning

confidence: 97%

Engaging in Dialogue about an Agent’s Norms and Behaviors

Kasenberg¹,

Roque²,

Thielstrom³

et al. 2019

Proceedings of the 1st Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence (NL4XAI 2019

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Contributionmentioning

confidence: 97%

Engaging in Dialogue about an Agent’s Norms and Behaviors

Kasenberg¹,

Roque²,

Thielstrom³

et al. 2019

Proceedings of the 1st Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence (NL4XAI 2019

Self Cite

View full text Add to dashboard Cite

show abstract

“…Manome et al (2018) is one of the few logic-to-text approaches using a sequence-to-sequence framework. Kasenberg et al (2019) generate explanations from a well-defined logical formalism in the context of human-robot dialogue.…”

Section: Related Workmentioning

confidence: 99%

Enhancing and Evaluating the Grammatical Framework Approach to Logic-to-Text Generation

Calò¹,

Werf²,

Gatt³

et al. 2022

Proceedings of the 2nd Workshop on Natural Language Generation, Evaluation, and Metrics (GEM)

View full text Add to dashboard Cite

Logic-to-text generation is an important yet underrepresented area of natural language generation (NLG). In particular, most previous works on this topic lack sound evaluation. We address this limitation by building and evaluating a system that generates high-quality English text given a first-order logic (FOL) formula as input. We start by analyzing the performance of Ranta (2011)'s system. Based on this analysis, we develop an extended version of the system, which we name LOLA, that performs formula simplification based on logical equivalences and syntactic transformations. We carry out an extensive evaluation of LOLA using standard automatic metrics and human evaluation. We compare the results against a baseline and Ranta (2011)'s system. The results show that LOLA outperforms the other two systems in most aspects.https://gitlab.nl4xai.eu/eduardo. calo/LoLa

show abstract

“…However, they do not consider how to generate the different plan traces, or how they can be used to answer specific questions. Kasenberg et al (2019) focus on justifying an agent's behaviour based on deterministic Markov decision problems. They construct explanations for the behaviour of an agent governed by temporal logic rules and answer questions including contrastive "why" queries.…”

Section: Contrastive Explanations Of Plansmentioning

confidence: 99%

Contrastive Explanations of Plans through Model Restrictions

Krarup¹,

Krivić

Magazzeni

et al. 2021

jair

View full text Add to dashboard Cite

In automated planning, the need for explanations arises when there is a mismatch between a proposed plan and the user’s expectation. We frame Explainable AI Planning as an iterative plan exploration process, in which the user asks a succession of contrastive questions that lead to the generation and solution of hypothetical planning problems that are restrictions of the original problem. The object of the exploration is for the user to understand the constraints that govern the original plan and, ultimately, to arrive at a satisfactory plan. We present the results of a user study that demonstrates that when users ask questions about plans, those questions are usually contrastive, i.e. “why A rather than B?”. We use the data from this study to construct a taxonomy of user questions that often arise during plan exploration. Our approach to iterative plan exploration is a process of successive model restriction. Each contrastive user question imposes a set of constraints on the planning problem, leading to the construction of a new hypothetical planning problem as a restriction of the original. Solving this restricted problem results in a plan that can be compared with the original plan, admitting a contrastive explanation. We formally define model-based compilations in PDDL2.1 for each type of constraint derived from a contrastive user question in the taxonomy, and empirically evaluate the compilations in terms of computational complexity. The compilations were implemented as part of an explanation framework supporting iterative model restriction. We demonstrate its benefits in a second user study.

show abstract

Generating justifications for norm-related agent decisions

Cited by 14 publications

References 10 publications

Engaging in Dialogue about an Agent’s Norms and Behaviors

Engaging in Dialogue about an Agent’s Norms and Behaviors

Enhancing and Evaluating the Grammatical Framework Approach to Logic-to-Text Generation

Contrastive Explanations of Plans through Model Restrictions

Contact Info

Product

Resources

About