arXiv subjects

Jonas Ulmen

Publications and source records attributed to Jonas Ulmen.

7 recordsLinked to original sources

arXiv2026

Component-Aware Pruning Framework for Neural Network Controllers via Gradient-Based Importance Estimation

The transition from monolithic to multi-component neural architectures in advanced neural network controllers poses substantial challenges due to the high computational complexity of the latter. Conventional model compression techniques for complexity reduction, such as structured pruning based on norm-based metrics to estimate the relative importance of distinct parameter groups, often fail to capture functional significance. This paper introduces a component-aware pruning framework that utilizes gradient information to compute three distinct importance metrics during training: Gradient Accumulation, Fisher Information, and Bayesian Uncertainty. Experimental results with an autoencoder and a TD-MPC agent demonstrate that the proposed framework reveals critical structural dependencies and dynamic shifts in importance that static heuristics often miss, supporting more informed compression decisions.

cs.LG↗

arXiv2026

A Latent Space Framework for Modeling Transient Engine Emissions Using Joint Embedding Predictive Architectures

Accurately modeling and controlling vehicle exhaust emissions during transient events, such as rapid acceleration, is critical for meeting environmental regulations and optimizing powertrains. Conventional data-driven methods, such as Multilayer Perceptrons (MLPs) and Long Short-Term Memory (LSTM) networks, improve upon phenomenological models but often struggle with the complex nonlinear dynamics of emission formation. These monolithic architectures are sensitive to dataset variability and typically require deep, computationally expensive structures to perform well, limiting their practical utility. This paper introduces a novel approach that overcomes these limitations by modeling emission dynamics within a structured latent space. Leveraging a Joint Embedding Predictive Architecture (JEPA), the proposed framework learns from a rich dataset that combines real-world Portable Emission Measurement System (PEMS) data with high-frequency hardware-in-the-loop measurements. The model abstracts away irrelevant noise, encoding only the key factors governing emission behavior into a compact, robust representation. This results in superior data efficiency and predictive accuracy across diverse transient regimes, significantly outperforming high-performing LSTM baselines in generalization. To ensure suitability for real-world deployment, the JEPA framework is structured to support pruning and post-training quantization. This strategy drastically reduces the computational footprint, minimizing inference time and memory demand with negligible accuracy loss. The result is a highly efficient model ideal for on-board implementation of advanced strategies, such as model predictive control or model-based reinforcement learning, in conventional and hybrid powertrains. These findings offer a clear pathway toward more robust emission control systems for next-generation vehicles.

eess.SY↗

arXiv2025

Learning State-Space Models of Dynamic Systems from Arbitrary Data using Joint Embedding Predictive Architectures

With the advent of Joint Embedding Predictive Architectures (JEPAs), which appear to be more capable than reconstruction-based methods, this paper introduces a novel technique for creating world models using continuous-time dynamic systems from arbitrary observation data. The proposed method integrates sequence embeddings with neural ordinary differential equations (neural ODEs). It employs loss functions that enforce contractive embeddings and Lipschitz constants in state transitions to construct a well-organized latent state space. The approach's effectiveness is demonstrated through the generation of structured latent state-space models for a simple pendulum system using only image data. This opens up a new technique for developing more general control algorithms and estimation techniques with broad applications in robotics.

cs.LG↗

arXiv2025

COMponent-Aware Pruning for Accelerated Control Tasks in Latent Space Models

The rapid growth of resource-constrained mobile platforms, including mobile robots, wearable systems, and Internet-of-Things devices, has increased the demand for computationally efficient neural network controllers (NNCs) that can operate within strict hardware limitations. While deep neural networks (DNNs) demonstrate superior performance in control applications, their substantial computational complexity and memory requirements present significant barriers to practical deployment on edge devices. This paper introduces a comprehensive model compression methodology that leverages component-aware structured pruning to determine the optimal pruning magnitude for each pruning group, ensuring a balance between compression and stability for NNC deployment. Our approach is rigorously evaluated on Temporal Difference Model Predictive Control (TD-MPC), a state-of-the-art model-based reinforcement learning algorithm, with a systematic integration of mathematical stability guarantee properties, specifically Lyapunov criteria. The key contribution of this work lies in providing a principled framework for determining the theoretical limits of model compression while preserving controller stability. Experimental validation demonstrates that our methodology successfully reduces model complexity while maintaining requisite control performance and stability characteristics. Furthermore, our approach establishes a quantitative boundary for safe compression ratios, enabling practitioners to systematically determine the maximum permissible model reduction before violating critical stability properties, thereby facilitating the confident deployment of compressed NNCs in resource-limited environments.

cs.RO↗

arXiv2025

Application-Specific Component-Aware Structured Pruning of Deep Neural Networks in Control via Soft Coefficient Optimization

Deep neural networks (DNNs) offer significant flexibility and robust performance. This makes them ideal for building not only system models but also advanced neural network controllers (NNCs). However, their high complexity and computational needs often limit their use. Various model compression strategies have been developed over the past few decades to address these issues. These strategies are effective for general DNNs but do not directly apply to NNCs. NNCs need both size reduction and the retention of key application-specific performance features. In structured pruning, which removes groups of related elements, standard importance metrics often fail to protect these critical characteristics. In this paper, we introduce a novel framework for calculating importance metrics in pruning groups. This framework not only shrinks the model size but also considers various application-specific constraints. To find the best pruning coefficient for each group, we evaluate two approaches. The first approach involves simple exploration through grid search. The second utilizes gradient descent optimization, aiming to balance compression and task performance. We test our method in two use cases: one on an MNIST autoencoder and the other on a Temporal Difference Model Predictive Control (TDMPC) agent. Results show that the method effectively maintains application-relevant performance while achieving a significant reduction in model size.

cs.LG↗

arXiv2025

Enhanced Pruning Strategy for Multi-Component Neural Architectures Using Component-Aware Graph Analysis

Deep neural networks (DNNs) deliver outstanding performance, but their complexity often prohibits deployment in resource-constrained settings. Comprehensive structured pruning frameworks based on parameter dependency analysis reduce model size with specific regard to computational performance. When applying them to Multi-Component Neural Architectures (MCNAs), they risk network integrity by removing large parameter groups. We introduce a component-aware pruning strategy, extending dependency graphs to isolate individual components and inter-component flows. This creates smaller, targeted pruning groups that conserve functional integrity. Demonstrated effectively on a control task, our approach achieves greater sparsity and reduced performance degradation, opening a path for optimizing complex, multi-component DNNs efficiently.

cs.LG↗

arXiv2024

State Derivative Normalization for Continuous-Time Deep Neural Networks

The importance of proper data normalization for deep neural networks is well known. However, in continuous-time state-space model estimation, it has been observed that improper normalization of either the hidden state or hidden state derivative of the model estimate, or even of the time interval can lead to numerical and optimization challenges with deep learning based methods. This results in a reduced model quality. In this contribution, we show that these three normalization tasks are inherently coupled. Due to the existence of this coupling, we propose a solution to all three normalization challenges by introducing a normalization constant at the state derivative level. We show that the appropriate choice of the normalization constant is related to the dynamics of the to-be-identified system and we derive multiple methods of obtaining an effective normalization constant. We compare and discuss all the normalization strategies on a benchmark problem based on experimental data from a cascaded tanks system and compare our results with other methods of the identification literature.

eess.SY↗