Autonomous capability assessment of sequential decision-making systems in stochastic settings
It is essential for users to understand what their AI systems can and can't do in order to use
them safely. However, the problem of enabling users to assess AI systems with sequential …
them safely. However, the problem of enabling users to assess AI systems with sequential …
A neurosymbolic cognitive architecture framework for handling novelties in open worlds
Abstract “Open world” environments are those in which novel objects, agents, events, and
more can appear and contradict previous understandings of the environment. This runs …
more can appear and contradict previous understandings of the environment. This runs …
JEDAI: A system for skill-aligned explainable robot planning
This paper presents JEDAI, an AI system designed for outreach and educational efforts
aimed at non-AI experts. JEDAI features a novel synthesis of research ideas from integrated …
aimed at non-AI experts. JEDAI features a novel synthesis of research ideas from integrated …
A domain-independent agent architecture for adaptive operation in evolving open worlds
Abstract Model-based reasoning agents are ill-equipped to act in novel situations in which
their model of the environment no longer sufficiently represents the world. We propose …
their model of the environment no longer sufficiently represents the world. We propose …
From Reals to Logic and Back: Inventing Symbolic Vocabularies, Actions and Models for Planning from Raw Data
Hand-crafted, logic-based state and action representations have been widely used to
overcome the intractable computational complexity of long-horizon robot planning problems …
overcome the intractable computational complexity of long-horizon robot planning problems …
Discovering user-interpretable capabilities of black-box planning agents
Several approaches have been developed for answering users' specific questions about AI
behavior and for assessing their core functionality in terms of primitive executable actions …
behavior and for assessing their core functionality in terms of primitive executable actions …
Autonomous capability assessment of black-box sequential decision-making systems
It is essential for users to understand what their AI systems can and can't do in order to use
them safely. However, the problem of enabling users to assess AI systems with evolving …
them safely. However, the problem of enabling users to assess AI systems with evolving …
Learning AI-System Capabilities under Stochasticity
Learning interpretable generalizable models of sequential decision-making agents is
essential for user-driven assessment as well as for continual agent-design processes in …
essential for user-driven assessment as well as for continual agent-design processes in …
Learning Generalizable and Composable Abstractions for Transfer in Reinforcement Learning
RK Nayyar - Proceedings of the AAAI Conference on Artificial …, 2024 - ojs.aaai.org
Reinforcement Learning (RL) in complex environments presents many challenges: agents
require learning concise representations of both environments and behaviors for efficient …
require learning concise representations of both environments and behaviors for efficient …
Data Efficient Paradigms for Personalized Assessment of Black-Box Taskable AI Systems
P Verma - Proceedings of the AAAI Conference on Artificial …, 2024 - ojs.aaai.org
The vast diversity of internal designs of taskable black-box AI systems and their nuanced
zones of safe functionality make it difficult for a layperson to use them without unintended …
zones of safe functionality make it difficult for a layperson to use them without unintended …