Benchmarks for automated commonsense reasoning: A survey
E Davis - ACM Computing Surveys, 2023 - dl.acm.org
More than one hundred benchmarks have been developed to test the commonsense
knowledge and commonsense reasoning abilities of artificial intelligence (AI) systems …
knowledge and commonsense reasoning abilities of artificial intelligence (AI) systems …
A survey of zero-shot generalisation in deep reinforcement learning
The study of zero-shot generalisation (ZSG) in deep Reinforcement Learning (RL) aims to
produce RL algorithms whose policies generalise well to novel unseen situations at …
produce RL algorithms whose policies generalise well to novel unseen situations at …
The signature-testing approach to mapping biological and artificial intelligences
Making inferences from behaviour to cognition is problematic due to a many-to-one mapping
problem, in which any one behaviour can be generated by multiple possible cognitive …
problem, in which any one behaviour can be generated by multiple possible cognitive …
Abstraction for deep reinforcement learning
M Shanahan, M Mitchell - arXiv preprint arXiv:2202.05839, 2022 - arxiv.org
We characterise the problem of abstraction in the context of deep reinforcement learning.
Various well established approaches to analogical reasoning and associative memory might …
Various well established approaches to analogical reasoning and associative memory might …
Synergistic information supports modality integration and flexible learning in neural networks solving multiple tasks
Striking progress has been made in understanding cognition by analyzing how the brain is
engaged in different modes of information processing. For instance, so-called synergistic …
engaged in different modes of information processing. For instance, so-called synergistic …
Detect, understand, act: A neuro-symbolic hierarchical reinforcement learning framework
In this paper we introduce Detect, Understand, Act (DUA), a neuro-symbolic reinforcement
learning framework. The Detect component is composed of a traditional computer vision …
learning framework. The Detect component is composed of a traditional computer vision …
Building thinking machines by solving animal cognition tasks
M Crosby - Minds and Machines, 2020 - Springer
Abstract In 'Computing Machinery and Intelligence', Turing, sceptical of the question 'Can
machines think?', quickly replaces it with an experimentally verifiable test: the imitation …
machines think?', quickly replaces it with an experimentally verifiable test: the imitation …
General intelligence disentangled via a generality metric for natural and artificial intelligence
Success in all sorts of situations is the most classical interpretation of general intelligence.
Under limited resources, however, the capability of an agent must necessarily be limited too …
Under limited resources, however, the capability of an agent must necessarily be limited too …
Evaluating AI evaluation: Perils and prospects
J Burden - arXiv preprint arXiv:2407.09221, 2024 - arxiv.org
As AI systems appear to exhibit ever-increasing capability and generality, assessing their
true potential and safety becomes paramount. This paper contends that the prevalent …
true potential and safety becomes paramount. This paper contends that the prevalent …
Memory gym: Partially observable challenges to memory-based agents
M Pleines, M Pallasch, F Zimmer… - … conference on learning …, 2023 - openreview.net
Memory Gym is a novel benchmark for challenging Deep Reinforcement Learning agents to
memorize events across long sequences, be robust to noise, and generalize. It consists of …
memorize events across long sequences, be robust to noise, and generalize. It consists of …