# KAN-Dreamer: Benchmarking Kolmogorov-Arnold Networks as Function Approximators in World Models

arXiv: https://arxiv.org/abs/2512.07437 · HTML: https://mlsanity.com/paper/2512.07437/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Latent dynamics models and model-based RL, one of 11 directions in World models.
- Builds on 21 of its 31 references that are in the map.
- First posted on arXiv 8 Dec 2025.

## Placed in

- World models / [Latent dynamics models and model-based RL](https://mlsanity.com/wm/latent-dynamics-models-and-model-based-rl/)
- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 21 of 21)

- [Dream to Control: Learning Behaviors by Latent Imagination](https://mlsanity.com/paper/1912.01603/) (2019)
- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Learning Latent Dynamics for Planning from Pixels](https://mlsanity.com/paper/1811.04551/) (2018)
- [Planning-oriented Autonomous Driving](https://mlsanity.com/paper/2212.10156/) (2022)
- [Mastering Atari with Discrete World Models](https://mlsanity.com/paper/2010.02193/) (2020)
- [DeepMind Control Suite](https://mlsanity.com/paper/1801.00690/) (2018)
- [Genie: Generative Interactive Environments](https://mlsanity.com/paper/2402.15391/) (2024)
- [KAN: Kolmogorov-Arnold Networks](https://mlsanity.com/paper/2404.19756/) (2024)
- [GAIA-1: A Generative World Model for Autonomous Driving](https://mlsanity.com/paper/2309.17080/) (2023)
- [DayDreamer: World Models for Physical Robot Learning](https://mlsanity.com/paper/2206.14176/) (2022)
- [Training Agents Inside of Scalable World Models](https://mlsanity.com/paper/2509.24527/) (2025)
- [Kolmogorov-Arnold Networks are Radial Basis Function Networks](https://mlsanity.com/paper/2405.06721/) (2024)
- [KAN 2.0: Kolmogorov-Arnold Networks Meet Science](https://mlsanity.com/paper/2408.10205/) (2024)
- [Kolmogorov-Arnold Networks (KANs) for Time Series Analysis](https://mlsanity.com/paper/2405.08790/) (2024)
- [The Kolmogorov-Arnold representation theorem revisited](https://mlsanity.com/paper/2007.15884/) (2020)
- [Kolmogorov Arnold Informed neural network: A physics-informed deep learning framework for solving forward and inverse problems based on Kolmogorov Arnold Networks](https://mlsanity.com/paper/2406.11045/) (2024)
- [KKANs: Kurkova-Kolmogorov-Arnold Networks and Their Learning Dynamics](https://mlsanity.com/paper/2412.16738/) (2024)
- [asKAN: Active Subspace embedded Kolmogorov-Arnold Network](https://mlsanity.com/paper/2504.04669/) (2025)
- [KAE: Kolmogorov-Arnold Auto-Encoder for Representation Learning](https://mlsanity.com/paper/2501.00420/) (2024)
- [Interpretable Reinforcement Learning for Load Balancing using Kolmogorov-Arnold Networks](https://mlsanity.com/paper/2505.14459/) (2025)
- [Sinusoidal Approximation Theorem for Kolmogorov-Arnold Networks](https://mlsanity.com/paper/2508.00247/) (2025)

## Built on by (most cited first, 1 of 1)

- [Sample-Efficiency of Kolmogorov-Arnold Networks](https://mlsanity.com/paper/2610.10627/) (2026)

## Abstract (arXiv)

DreamerV3 is a state-of-the-art online model-based reinforcement learning (MBRL) algorithm known for remarkable sample efficiency. Concurrently, Kolmogorov-Arnold Networks (KANs) have emerged as a promising alternative to Multi-Layer Perceptrons (MLPs), offering superior parameter efficiency and interpretability. To mitigate KANs' computational overhead, variants like FastKAN leverage Radial Basis Functions (RBFs) to accelerate inference. In this work, we investigate integrating KAN architectures into the DreamerV3 framework. We introduce KAN-Dreamer, replacing specific MLP and convolutional components of DreamerV3 with KAN and FastKAN layers. To ensure efficiency within the JAX-based World Model, we implement a tailored, fully vectorized version with simplified grid management. We structure our investigation into three subsystems: Visual Perception, Latent Prediction, and Behavior Learning. Empirical evaluations on the DeepMind Control Suite (walker_walk) analyze sample efficiency, training time, and asymptotic performance. Experimental results demonstrate that utilizing our adapted FastKAN as a drop-in replacement for the Reward and Continue predictors yields performance on par with the original MLP-based architecture, maintaining parity in both sample efficiency and training speed. This report serves as a preliminary study for future developments in KAN-based world models.