Bayesian Belief Updating in Neural Networks
An analysis of how deep learning architectures approximate Bayesian inference, including limitations in priors selection and posterior calibration.
The interdisciplinary study of how knowledge is represented, computed, and acquired by artificial systems and cognitive architectures. Bridging philosophy of mind, computer science, and cognitive psychology to formalize belief updating, justification, and truth in algorithmic contexts.
An analysis of how deep learning architectures approximate Bayesian inference, including limitations in priors selection and posterior calibration.
Exploring how classic philosophical counterexamples to justified true belief manifest in AI-generated content and automated fact-checking pipelines.
Interactive knowledge graph mapping cognitive linguistics concepts to computational semantic networks and NLP vector space models.
Formalizing computational approaches to detecting misinformation, including trust metrics, source provenance tracking, and adversarial robustness.
How modern embodied AI systems challenge traditional internalist epistemology, proposing frameworks for distributed knowledge states.
A survey of modal logics used to model knowledge, belief, and common knowledge in decentralized AI networks and blockchain consensus.