Abstract:Artificial Neural Networks (ANNs) have significantly advanced various fields by effectively recognizing patterns and solving complex problems. Despite these advancements, their interpretability remains a critical challenge, especially in applications where transparency and accountability are essential. To address this, explainable AI (XAI) has made progress in demystifying ANNs, yet interpretability alone is often insufficient. In certain applications, model predictions must align with expert-imposed requirements, sometimes exemplified by partial monotonicity constraints. While monotonic approaches are found in the literature for traditional Multi-layer Perceptrons (MLPs), they still face difficulties in achieving both interpretability and certified partial monotonicity. Recently, the Kolmogorov-Arnold Network (KAN) architecture, based on learnable activation functions parametrized as splines, has been proposed as a more interpretable alternative to MLPs. Building on this, we introduce a novel ANN architecture called MonoKAN, which is based on the KAN architecture and achieves certified partial monotonicity while enhancing interpretability. To achieve this, we employ cubic Hermite splines, which guarantee monotonicity through a set of straightforward conditions. Additionally, by using positive weights in the linear combinations of these splines, we ensure that the network preserves the monotonic relationships between input and output. Our experiments demonstrate that MonoKAN not only enhances interpretability but also improves predictive performance across the majority of benchmarks, outperforming state-of-the-art monotonic MLP approaches.

Kolmogorov-Arnold Network for Online Reinforcement Learning

KAN: Kolmogorov-Arnold Networks

KAN v.s. MLP for Offline Reinforcement Learning

Activation Space Selectable Kolmogorov-Arnold Networks

On the expressiveness and spectral bias of KANs

P1-KAN an effective Kolmogorov Arnold Network for function approximation

Kolmogorov-Arnold Networks (KAN) for Time Series Classification and Robust Analysis

Kolmogorov-Arnold Networks in Low-Data Regimes: A Comparative Study with Multilayer Perceptrons

Kolmogorov-Arnold Networks for Genomic Tasks

Leveraging KANs For Enhanced Deep Koopman Operator Discovery

On Training of Kolmogorov-Arnold Networks

A Benchmarking Study of Kolmogorov-Arnold Networks on Tabular Data

Kolmogorov-Arnold Network Autoencoders

A comprehensive and FAIR comparison between MLP and KAN representations for differential equations and operator networks

A Survey on Kolmogorov-Arnold Network

On the Robustness of Kolmogorov-Arnold Networks: An Adversarial Perspective

KAN-ODEs: Kolmogorov-Arnold Network Ordinary Differential Equations for Learning Dynamical Systems and Hidden Physics

MonoKAN: Certified Monotonic Kolmogorov-Arnold Network

Demonstrating the Efficacy of Kolmogorov-Arnold Networks in Vision Tasks

Reimagining Linear Probing: Kolmogorov-Arnold Networks in Transfer Learning