arXiv subjects

Joerg Reichardt

Publications and source records attributed to Joerg Reichardt.

At least 19 recordsLinked to original sources

arXiv2026

Active Interaction-Aware Model Predictive Path Integral via Ego-Conditioned Generative Predictions

Dense traffic is inherently interactive. The ego vehicle and surrounding agents continuously influence each other's reactions, making "what-if" reasoning essential for safe and efficient driving. To enable such an active interaction-aware behavior, we propose a planning framework that integrates an ego-conditioned generative autoregressive prediction model within Model Predictive Path Integral (MPPI) control. The generative prediction model outputs stochastic, multi-modal predictions of surrounding agents conditioned on each of the ego's considered future actions. A nested sampling scheme enables tractable evaluation of expected cost and collision risk under the induced distribution. This formulation allows the ego to actively probe how different candidate actions shape the interaction outcomes and to identify actions that reduce ambiguity in uncertain interactions. Closed-loop simulations demonstrate improved safety and efficiency compared to conventional predict-then-plan and passive interaction-aware approaches.

cs.RO↗

arXiv2025

EP-Diffuser: An Efficient Diffusion Model for Traffic Scene Generation and Prediction via Polynomial Representations

As the prediction horizon increases, predicting the future evolution of traffic scenes becomes increasingly difficult due to the multi-modal nature of agent motion. Most state-of-the-art (SotA) prediction models primarily focus on forecasting the most likely future. However, for the safe operation of autonomous vehicles, it is equally important to cover the distribution for plausible motion alternatives. To address this, we introduce EP-Diffuser, a novel parameter-efficient diffusion-based generative model designed to capture the distribution of possible traffic scene evolutions. Conditioned on road layout and agent history, our model acts as a predictor and generates diverse, plausible scene continuations. We benchmark EP-Diffuser against two SotA models in terms of accuracy and plausibility of predictions on the Argoverse 2 dataset. Despite its significantly smaller model size, our approach achieves both highly accurate and plausible traffic scene predictions. We further evaluate model generalization ability in an out-of-distribution (OoD) test setting using Waymo Open dataset and show superior robustness of our approach. The code and model checkpoints are available at: https://github.com/continental/EP-Diffuser.

cs.CV↗

arXiv2025

An Empirical Bayes Analysis of Object Trajectory Representation Models

Linear trajectory models provide mathematical advantages to autonomous driving applications such as motion prediction. However, linear models' expressive power and bias for real-world trajectories have not been thoroughly analyzed. We present an in-depth empirical analysis of the trade-off between model complexity and fit error in modelling object trajectories. We analyze vehicle, cyclist, and pedestrian trajectories. Our methodology estimates observation noise and prior distributions over model parameters from several large-scale datasets. Incorporating these priors can then regularize prediction models. Our results show that linear models do represent real-world trajectories with high fidelity at very moderate model complexity. This suggests the feasibility of using linear trajectory models in future motion prediction systems with inherent mathematical advantages.

cs.LG↗

arXiv2025

Reachability-Based Contingency Planning against Multi-Modal Predictions with Branch MPC

This paper presents a novel contingency planning framework that integrates learning-based multi-modal predictions of traffic participants into Branch Model Predictive Control (MPC). Leveraging reachability analysis, we address the computational challenges associated with Branch MPC by organizing the multitude of predictions into driving corridors. Analyzing the overlap between these corridors, their number can be reduced through pruning and clustering while ensuring safety since all prediction modes are preserved. These processed corridors directly correspond to the distinct branches of the scenario tree and provide an efficient constraint representation for the Branch MPC. We further utilize the reachability for determining maximum feasible decision postponing times, ensuring that branching decisions remain executable. Qualitative and quantitative evaluations demonstrate significantly reduced computational complexity and enhanced safety and comfort.

eess.SY↗

arXiv2025

Beyond In-Distribution Performance: A Cross-Dataset Study of Trajectory Prediction Robustness

We study the Out-of-Distribution (OoD) generalization ability of three SotA trajectory prediction models with comparable In-Distribution (ID) performance but different model designs. We investigate the influence of inductive bias, size of training data and data augmentation strategy by training the models on Argoverse 2 (A2) and testing on Waymo Open Motion (WO) and vice versa. We find that the smallest model with highest inductive bias exhibits the best OoD generalization across different augmentation strategies when trained on the smaller A2 dataset and tested on the large WO dataset. In the converse setting, training all models on the larger WO dataset and testing on the smaller A2 dataset, we find that all models generalize poorly, even though the model with the highest inductive bias still exhibits the best generalization ability. We discuss possible reasons for this surprising finding and draw conclusions about the design and test of trajectory prediction models and benchmarks.

cs.LG↗

arXiv2025

Improving Out-of-Distribution Generalization of Trajectory Prediction for Autonomous Driving via Polynomial Representations

Robustness against Out-of-Distribution (OoD) samples is a key performance indicator of a trajectory prediction model. However, the development and ranking of state-of-the-art (SotA) models are driven by their In-Distribution (ID) performance on individual competition datasets. We present an OoD testing protocol that homogenizes datasets and prediction tasks across two large-scale motion datasets. We introduce a novel prediction algorithm based on polynomial representations for agent trajectory and road geometry on both the input and output sides of the model. With a much smaller model size, training effort, and inference time, we reach near SotA performance for ID testing and significantly improve robustness in OoD testing. Within our OoD testing protocol, we further study two augmentation strategies of SotA models and their effects on model generalization. Highlighting the contrast between ID and OoD performance, we suggest adding OoD testing to the evaluation criteria of trajectory prediction models.

cs.LG↗

arXiv2024

Motion Planning under Uncertainty: Integrating Learning-Based Multi-Modal Predictors into Branch Model Predictive Control

In complex traffic environments, autonomous vehicles face multi-modal uncertainty about other agents' future behavior. To address this, recent advancements in learningbased motion predictors output multi-modal predictions. We present our novel framework that leverages Branch Model Predictive Control(BMPC) to account for these predictions. The framework includes an online scenario-selection process guided by topology and collision risk criteria. This efficiently selects a minimal set of predictions, rendering the BMPC realtime capable. Additionally, we introduce an adaptive decision postponing strategy that delays the planner's commitment to a single scenario until the uncertainty is resolved. Our comprehensive evaluations in traffic intersection and random highway merging scenarios demonstrate enhanced comfort and safety through our method.

cs.RO↗

arXiv2023

Learning-Aided Warmstart of Model Predictive Control in Uncertain Fast-Changing Traffic

Model Predictive Control lacks the ability to escape local minima in nonconvex problems. Furthermore, in fast-changing, uncertain environments, the conventional warmstart, using the optimal trajectory from the last timestep, often falls short of providing an adequately close initial guess for the current optimal trajectory. This can potentially result in convergence failures and safety issues. Therefore, this paper proposes a framework for learning-aided warmstarts of Model Predictive Control algorithms. Our method leverages a neural network based multimodal predictor to generate multiple trajectory proposals for the autonomous vehicle, which are further refined by a sampling-based technique. This combined approach enables us to identify multiple distinct local minima and provide an improved initial guess. We validate our approach with Monte Carlo simulations of traffic scenarios.

eess.SY↗

arXiv2015

Motifs in Triadic Random Graphs based on Steiner Triple Systems

Conventionally, pairwise relationships between nodes are considered to be the fundamental building blocks of complex networks. However, over the last decade the overabundance of certain sub-network patterns, so called motifs, has attracted high attention. It has been hypothesized, these motifs, instead of links, serve as the building blocks of network structures. Although the relation between a network's topology and the general properties of the system, such as its function, its robustness against perturbations, or its efficiency in spreading information is the central theme of network science, there is still a lack of sound generative models needed for testing the functional role of subgraph motifs. Our work aims to overcome this limitation. We employ the framework of exponential random graphs (ERGMs) to define novel models based on triadic substructures. The fact that only a small portion of triads can actually be set independently poses a challenge for the formulation of such models. To overcome this obstacle we use Steiner Triple Systems (STS). These are partitions of sets of nodes into pair-disjoint triads, which thus can be specified independently. Combining the concepts of ERGMs and STS, we suggest novel generative models capable of generating ensembles of networks with non-trivial triadic Z-score profiles. Further, we discover inevitable correlations between the abundance of triad patterns, which occur solely for statistical reasons and need to be taken into account when discussing the functional implications of motif statistics. Moreover, we calculate the degree distributions of our triadic random graphs analytically.

physics.soc-ph↗

arXiv2015

Node-Specific Triad Pattern Mining for Complex-Network Analysis

The mining of graphs in terms of their local substructure is a well-established methodology to analyze networks. It was hypothesized that motifs - subgraph patterns which appear significantly more often than expected at random - play a key role for the ability of a system to perform its task. Yet the framework commonly used for motif-detection averages over the local environments of all nodes. Therefore, it remains unclear whether motifs are overrepresented in the whole system or only in certain regions. In this contribution, we overcome this limitation by mining node-specific triad patterns. For every vertex, the abundance of each triad pattern is considered only in triads it participates in. We investigate systems of various fields and find that motifs are distributed highly heterogeneously. In particular we focus on the feed-forward loop motif which has been alleged to play a key role in biological networks.

cs.SI↗

arXiv2011

Information Symmetries in Irreversible Processes

We study dynamical reversibility in stationary stochastic processes from an information theoretic perspective. Extending earlier work on the reversibility of Markov chains, we focus on finitary processes with arbitrarily long conditional correlations. In particular, we examine stationary processes represented or generated by edge-emitting, finite-state hidden Markov models. Surprisingly, we find pervasive temporal asymmetries in the statistics of such stationary processes with the consequence that the computational resources necessary to generate a process in the forward and reverse temporal directions are generally not the same. In fact, an exhaustive survey indicates that most stationary processes are irreversible. We study the ensuing relations between model topology in different representations, the process's statistical properties, and its reversibility in detail. A process's temporal asymmetry is efficiently captured using two canonical unifilar representations of the generating model, the forward-time and reverse-time epsilon-machines. We analyze example irreversible processes whose epsilon-machine presentations change size under time reversal, including one which has a finite number of recurrent causal states in one direction, but an infinite number in the opposite. From the forward-time and reverse-time epsilon-machines, we are able to construct a symmetrized, but nonunifilar, generator of a process---the bidirectional machine. Using the bidirectional machine, we show how to directly calculate a process's fundamental information properties, many of which are otherwise only poorly approximated via process samples. The tools we introduce and the insights we offer provide a better understanding of the many facets of reversibility and irreversibility in stochastic processes.

cond-mat.stat-mech↗

arXiv2010

The interplay of microscopic and mesoscopic structure in complex networks

Not all nodes in a network are created equal. Differences and similarities exist at both individual node and group levels. Disentangling single node from group properties is crucial for network modeling and structural inference. Based on unbiased generative probabilistic exponential random graph models and employing distributive message passing techniques, we present an efficient algorithm that allows one to separate the contributions of individual nodes and groups of nodes to the network structure. This leads to improved detection accuracy of latent class structure in real world data sets compared to models that focus on group structure alone. Furthermore, the inclusion of hitherto neglected group specific effects in models used to assess the statistical significance of small subgraph (motif) distributions in networks may be sufficient to explain most of the observed statistics. We show the predictive power of such generative models in forecasting putative gene-disease associations in the Online Mendelian Inheritance in Man (OMIM) database. The approach is suitable for both directed and undirected uni-partite as well as for bipartite networks.

cond-mat.stat-mech↗

arXiv2008

Protein-Interaction-Networks: More than mere modules

Cellular function is widely believed to be organized in a modular fashion. On all scales and at all levels of complexity, relatively independent sub-units perform relatively independent sub-tasks of biological function. This functional modularity must be reflected in the topology of molecular networks. But how a functional module should be represented in an interaction network is an open question. In protein-interaction networks (PIN), one can identify a protein-complex as a module on a small scale, i.e. modules are understood as densely linked, resp. interacting, groups of proteins, that are only sparsely interacting with the rest of the network. In this contribution, we show that extrapolating this concept of cohesively linked clusters of proteins as modules to the scale of the entire PIN inevitable misses important and functionally relevant structure inherent in the network. As an alternative, we introduce a novel way of decomposing a network into functional roles and show that this represents network structure and function more efficiently. This finding should have a profound impact on all module assisted methods of protein function prediction and should shed new light on how functional modules can be represented in molecular interaction networks in general.

q-bio.MN↗

arXiv2007

(Un)detectable cluster structure in sparse networks

We study the problem of recovering a known cluster structure in a sparse network, also known as the planted partitioning problem, by means of statistical mechanics. We find a sharp transition from un-recoverable to recoverable structure as a function of the separation of the clusters. For multivariate data, such transitions have been observed frequently, but always as a function of the number of data points provided, i.e. given a large enough data set, two point clouds can always be recognized as different clusters, as long as their separation is non-zero. In contrast, for the sparse networks studied here, a cluster structure remains undetectable even in an infinitely large network if a critical separation is not exceeded. We give analytic formulas for this critical separation as a function of the degree distribution of the network and calculate the shape of the recoverability-transition. Our findings have implications for unsupervised learning and data-mining in relational data bases and provide bounds on the achievable performance of graph clustering algorithms.

cond-mat.dis-nn↗

arXiv2007

Role models for complex networks

We present a framework for automatically decomposing ("block-modeling") the functional classes of agents within a complex network. These classes are represented by the nodes of an image graph ("block model") depicting the main patterns of connectivity and thus functional roles in the network. Using a first principles approach, we derive a measure for the fit of a network to any given image graph allowing objective hypothesis testing. From the properties of an optimal fit, we derive how to find the best fitting image graph directly from the network and present a criterion to avoid overfitting. The method can handle both two-mode and one-mode data, directed and undirected as well as weighted networks and allows for different types of links to be dealt with simultaneously. It is non-parametric and computationally efficient. The concepts of structural equivalence and modularity are found as special cases of our approach. We apply our method to the world trade network and analyze the roles individual countries play in the global economy.

physics.soc-ph↗

arXiv2007

Partitioning and modularity of graphs with arbitrary degree distribution

We solve the graph bi-partitioning problem in dense graphs with arbitrary degree distribution using the replica method. We find the cut-size to scale universally with . In contrast, earlier results studying the problem in graphs with a Poissonian degree distribution had found a scaling with ^1/2 [Fu and Anderson, J. Phys. A: Math. Gen. 19, 1986]. The new results also generalize to the problem of q-partitioning. They can be used to find the expected modularity Q [Newman and Grivan, Phys. Rev. E, 69, 2004] of random graphs and allow for the assessment of statistical significance of the output of community detection algorithms.

cond-mat.dis-nn↗

arXiv2006

eBay users form stable groups of common interest

Market segmentation of an online auction site is studied by analyzing the users' bidding behavior. The distribution of user activity is investigated and a network of bidders connected by common interest in individual articles is constructed. The network's cluster structure corresponds to the main user groups according to common interest, exhibiting hierarchy and overlap. Key feature of the analysis is its independence of any similarity measure between the articles offered on eBay, as such a measure would only introduce bias in the analysis. Results are compared to null models based on random networks and clusters are validated and interpreted using the taxonomic classifications of eBay categories. We find clear-cut and coherent interest profiles for the bidders in each cluster. The interest profiles of bidder groups are compared to the classification of articles actually bought by these users during the time span 6-9 months after the initial grouping. The interest profiles discovered remain stable, indicating typical interest profiles in society. Our results show how network theory can be applied successfully to problems of market segmentation and sociological milieu studies with sparse, high dimensional data.

physics.soc-ph↗

arXiv2006

When are networks truly modular?

Study of the cluster- or community structure of complex networks makes an important contribution to the understanding of networks at a functional level. Despite the many efforts, no definition of community has been agreed on and important aspects such as the statistical significance and theoretical limits of community detection are not well understood. We show how the problem of community detection can be mapped onto finding the ground state of an infinite range spin glass. The ground state energy then corresponds directly to the quality of the partition. The network modularity Q previously defined by Girvan and Newman [Phys. Rev. E, 69, 026113 (2004)] turns out to be a special case of this spin glass energy. Through this spin glass analogy, we are able to give expectation values for the modularity of random graphs that can be used in the assessment of the statistical significance of real network clusterings. Further, it allows for assessing the theoretical limits of community detection.

cond-mat.dis-nn↗