A Neurocognitive Approach to Robotic Cause-Effect Reasoning During Learning

reportActive / Technical Report | Accesssion Number: AD1169707 | Open PDF

Abstract:

This project developed a purely neurocomputational cognitive architecture for robotic systems that are doing imitation learning (learning from demonstrations). The main results were: 1. implementation of a neural virtual machine; 2. implementation of a neural system that automatically creates explanations for a human demonstrators intentions/goals by using cause-effect reasoning; and 3. comparative analysis of human versus robot behavioral activities during imitation learning.

Author(s):
Gentili, Rodolphe ; Katz, Garrett ; Reggia, James
Author Organization(s):
MARYLAND UNIV COLLEGE PARK; SYRACUSE UNIV NY
Funding Organization(s):
OFFICE OF NAVAL RESEARCH , ARLINGTON , VA
Document Type:
Technical Report/Final Report
Publication Date:
2022 May 23
Pagination:
7

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution Code:
A - Approved For Public Release
Distribution Statement: Public Release

RECORD

Collection: TRECMS
Identifying Numbers
Grant Number(s):
N00014-19-1-2044
Subject Terms
Modernization Areas:
Autonomy
Communities of Interest:
No COI(s) Identified
Descriptor(s):
artificial intelligence, neural networks, reasoning, computer science, cognitive workload, machine learning, virtual machines, computer programming, lisp programming language, causal reasoning, cognitive systems engineering, computational neuroscience, demonstrations, abstracts, algorithms, computers, maryland
Keyword(s):
Robotic imitation learning, programmable neural networks, learning from demonstrations, Neurocognition, Robotic Cause-Effect Reasoning, nvm (neural virtual machines)
Subject Categories:
Mathematical and Computer Sciences
Creation Date:
2022 May 24
Update Date:
2022 Jul 13