Goal-Directed Procedure Learning

Hammer, Patrick; Lofthouse, Tony

doi:10.1007/978-3-319-97676-1_8

Cited by 9 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We decided to use NARS [8] to assist during planning activities. The efficiency of NARS to learn procedure knowledge has been proven using the mechanisms described in [7]. NARS provides a robust reasoning cycle when robots have to produce standalone actions and when it is crucial to find alternative solutions to the ones determined at design time.…”

Section: Discussionmentioning

confidence: 99%

Agents in dynamic contexts, a system for learning plans

Lanza

Hammer

Seidita

et al. 2020

Proceedings of the 35th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Agents in dynamic contexts, a system for learning plans

Lanza

Hammer

Seidita

et al. 2020

Proceedings of the 35th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…A prototype of the system, OpenNARS, has been implemented and tested in a variety of domains requiring, for example, Procedure Learning, diagnostics, Question Answering, and Anomaly Detection. [11]. In particular, the use cases tested include Procedure Learning tasks such as Pong, Real-Time Anomaly Detection (jaywalking and Pedestrian Danger) and Question-Answering in the Street Scene Dataset,as well as robot control in the TestChamber simulation developed for OpenNARS.…”

Section: Nars: Non Axiomatic Reasoning Systemmentioning

confidence: 99%

“…These ideas, originally proposed in [11], have been implemented in OpenNARS and are sufficient to explain the role of NARS in our integration with the Jason planner. A key property of NARS that our integration exploits is that NARS can revise and withdraw failed plans when the success rate becomes too low, and new procedural knowledge can be added to the plan library as AgentSpeak plans, ultimately allowing AgentSpeak to be used in a less constrained way for autonomous systems facing changing environments.…”

Section: Nars: Non Axiomatic Reasoning Systemmentioning

confidence: 99%

“…We chose to use NARS [5] to support the planning activities. The efficiency of NARS for procedural knowledge learning was demonstrated using the mechanisms described in [11]. NARS provides a robust reasoning cycle in situations where robots need to perform autonomous actions and where it is crucial to find alternative solutions to those defined at planning time.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Deliberative Agents in Dynamic Environments, Using Jason and NARS

Seidita¹,

Lanza²,

Hammer³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

This work explore the possibility to combine the Jason reasoning cycle with a Non-Axiomatic Reasoning System (NARS) to develop multi-agent systems that are able to reason, deliberate and plan when information about plans to be executed and goals to be pursued is missing or incomplete. The contribution of this work is a method for BDI agents to create high-level plans using an AGI (Artificial General Intelligence) system based on non-axiomatic logic.

show abstract

“…An ideal artificial cumulative learner, in our conceptualization, can therefore acquire knowledge and skills through both experience [20] and explicit teaching [3]. Goal-generality means that knowledge and goal(s) are not fused together (in particular situations and constraints) but can be re-purposed when task-and domain-related parameters change [9].…”

Section: Generalitymentioning

confidence: 99%

Cumulative Learning

Þórisson

Bieger

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

An important feature of human learning is the ability to continuously accept new information and unify it with existing knowledge, a process that proceeds largely automatically and without catastrophic side-effects. A generally intelligent machine (AGI) should be able to learn a wide range of tasks in a variety of environments. Knowledge acquisition in partially-known and dynamic task-environments cannot happen all-at-once, and AGI-aspiring systems must thus be capable of cumulative learning: efficiently making use of existing knowledge while learning new things, increasing the scope of ability and knowledge incrementally-without catastrophic forgetting or damaging existing skills. Many aspects of such learning have been addressed in artificial intelligence (AI) research, but relatively few examples of cumulative learning have been demonstrated to date and no generally accepted explicit definition exists of this category of learning. Here we provide a general definition of cumulative learning and describe how it relates to other concepts frequently used in the AI literature.

show abstract

Goal-Directed Procedure Learning

Cited by 9 publications

References 9 publications

Agents in dynamic contexts, a system for learning plans

Agents in dynamic contexts, a system for learning plans

Deliberative Agents in Dynamic Environments, Using Jason and NARS

Cumulative Learning

Contact Info

Product

Resources

About