Answers are grounded in this paper and cite their sources.
Fractional calculus provides a mathematical framework for representing memory, hereditary effects and long-range dependence. These properties have attracted increasing attention in artificial intelligence (AI), where learning systems must process historical information, optimize complex objectives and model nonlinear dynamics. This review summarizes the emerging integration of fractional calculus with machine learning, neural networks, optimization, physics-informed artificial intelligence, generative AI and applications in finance, commerce and management. Particular attention is given to fractional-order gradient methods, fractional neural networks, fractional physics-informed neural networks and the emerging use of fractional mechanisms in large language models. The review also discusses the potential of hybrid fractional-AI models for intelligent decision-making and identifies major challenges involving computational cost, fractional-order selection, convergence, interpretability and large-scale implementation. The review shows that fractional calculus should be viewed as a complementary mathematical tool for developing memory-aware and domain-informed AI rather than as a universal replacement for conventional learning methods.
No takes yet. Share an insight, caveat, or question.
Swarupkumar Bhalke (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: