KASPER: Kolmogorov Arnold Networks for Stock Prediction and Explainable Regimes

2025-07-25

AI Summary

Financial market forecasting faces dual challenges: inherent nonlinearity and regime-dependent dynamics, leading to poor generalization and limited interpretability in existing deep learning models. To address these issues, we propose KASPER—a novel framework that (i) introduces the Kolmogorov–Arnold superposition network for financial time-series modeling; (ii) integrates Gumbel-Softmax sampling for end-to-end latent market regime identification; and (iii) employs sparse spline activation functions to capture nonlinear price dynamics within each regime. Furthermore, Monte Carlo Shapley value estimation is leveraged to extract symbolic, human-readable decision rules, ensuring both transparency and predictive performance. Evaluated on real-world Yahoo Finance data, KASPER achieves R² = 0.89, Sharpe ratio = 12.02, and MSE = 0.0001—significantly outperforming state-of-the-art baselines—demonstrating a unique synergy of high accuracy, robustness, and intrinsic interpretability.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Reasoning under Uncertainty: Relational Probabilistic Models
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Forecasting in financial markets remains a significant challenge due to their nonlinear and regime-dependent dynamics. Traditional deep learning models, such as long short-term memory networks and multilayer perceptrons, often struggle to generalize across shifting market conditions, highlighting the need for a more adaptive and interpretable approach. To address this, we introduce Kolmogorov-Arnold networks for stock prediction and explainable regimes (KASPER), a novel framework that integrates regime detection, sparse spline-based function modeling, and symbolic rule extraction. The framework identifies hidden market conditions using a Gumbel-Softmax-based mechanism, enabling regime-specific forecasting. For each regime, it employs Kolmogorov-Arnold networks with sparse spline activations to capture intricate price behaviors while maintaining robustness. Interpretability is achieved through symbolic learning based on Monte Carlo Shapley values, which extracts human-readable rules tailored to each regime. Applied to real-world financial time series from Yahoo Finance, the model achieves an $R^2$ score of 0.89, a Sharpe Ratio of 12.02, and a mean squared error as low as 0.0001, outperforming existing methods. This research establishes a new direction for regime-aware, transparent, and robust forecasting in financial markets.

Problem

• Predicting stock prices in nonlinear, regime-dependent financial markets
• Improving model adaptability and interpretability across shifting market conditions
• Integrating regime detection with explainable, robust forecasting techniques

Innovation

• Gumbel-Softmax for regime detection
• Sparse spline Kolmogorov-Arnold networks
• Monte Carlo Shapley symbolic rules