Vít Musil

Subham Sekhar Sahoo, Anselm Paulus, Marin Vlastelica et al.

Embedding discrete solvers as differentiable layers has given modern deep learning architectures combinatorial expressivity and discrete reasoning capabilities. The derivative of these solvers is zero or undefined, therefore a meaningful replacement is crucial for effective gradient-based learning. Prior works rely on smoothing the solver with input perturbations, relaxing the solver to continuous problems, or interpolating the loss landscape with techniques that typically require additional solver calls, introduce extra hyper-parameters, or compromise performance. We propose a principled approach to exploit the geometry of the discrete solution space to treat the solver as a negative identity on the backward pass and further provide a theoretical justification. Our experiments demonstrate that such a straightforward hyper-parameter-free approach is able to compete with previous more complex methods on numerous experiments such as backpropagation through discrete samplers, deep graph matching, and image retrieval. Furthermore, we substitute the previously proposed problem-specific and label-dependent margin with a generic regularization procedure that prevents cost collapse and increases robustness.

Anselm Paulus, Georg Martius, Vít Musil

Embedding parameterized optimization problems as layers into machine learning architectures serves as a powerful inductive bias. Training such architectures with stochastic gradient descent requires care, as degenerate derivatives of the embedded optimization problem often render the gradients uninformative. We propose Lagrangian Proximal Gradient Descent (LPGD) a flexible framework for training architectures with embedded optimization layers that seamlessly integrates into automatic differentiation libraries. LPGD efficiently computes meaningful replacements of the degenerate optimization layer derivatives by re-running the forward solver oracle on a perturbed input. LPGD captures various previously proposed methods as special cases, while fostering deep links to traditional optimization methods. We theoretically analyze our method and demonstrate on historical and synthetic data that LPGD converges faster than gradient descent even in a differentiable setup.

Martin Krebs, Jan Obdržálek, Vít Musil et al.

Deep neural networks are starting to show their worth in critical applications such as assisted cancer diagnosis. However, for their outputs to get accepted in practice, the results they provide should be explainable in a way easily understood by pathologists. A well-known and widely used explanation technique is occlusion, which, however, can take a long time to compute, thus slowing the development and interaction with pathologists. In this work, we set out to find a faster replacement for occlusion in a successful system for detecting prostate cancer. Since there is no established framework for comparing the performance of various explanation methods, we first identified suitable comparison criteria and selected corresponding metrics. Based on the results, we were able to choose a different explanation method, which cut the previously required explanation time at least by a factor of 10, without any negative impact on the quality of outputs. This speedup enables rapid iteration in model development and debugging and brings us closer to adopting AI-assisted prostate cancer detection in clinical settings. We propose that our approach to finding the replacement for occlusion can be used to evaluate candidate methods in other related applications.

Michal Rolínek, Paul Swoboda, Dominik Zietlow et al.

Building on recent progress at the intersection of combinatorial optimization and deep learning, we propose an end-to-end trainable architecture for deep graph matching that contains unmodified combinatorial solvers. Using the presence of heavily optimized combinatorial solvers together with some improvements in architecture design, we advance state-of-the-art on deep graph matching benchmarks for keypoint correspondence. In addition, we highlight the conceptual advantages of incorporating solvers into deep learning architectures, such as the possibility of post-processing with a strong multi-graph matching solver or the indifference to changes in the training setting. Finally, we propose two new challenging experimental setups. The code is available at https://github.com/martius-lab/blackbox-deep-graph-matching

Michal Rolínek, Vít Musil, Anselm Paulus et al.

Rank-based metrics are some of the most widely used criteria for performance evaluation of computer vision models. Despite years of effort, direct optimization for these metrics remains a challenge due to their non-differentiable and non-decomposable nature. We present an efficient, theoretically sound, and general method for differentiating rank-based metrics with mini-batch gradient descent. In addition, we address optimization instability and sparsity of the supervision signal that both arise from using rank-based metrics as optimization targets. Resulting losses based on recall and Average Precision are applied to image retrieval and object detection tasks. We obtain performance that is competitive with state-of-the-art on standard image retrieval datasets and consistently improve performance of near state-of-the-art object detectors. The code is available at https://github.com/martius-lab/blackbox-backprop

Marin Vlastelica, Anselm Paulus, Vít Musil et al.

Achieving fusion of deep learning with combinatorial algorithms promises transformative changes to artificial intelligence. One possible approach is to introduce combinatorial building blocks into neural networks. Such end-to-end architectures have the potential to tackle combinatorial problems on raw input data such as ensuring global consistency in multi-object tracking or route planning on maps in robotics. In this work, we present a method that implements an efficient backward pass through blackbox implementations of combinatorial solvers with linear objective functions. We provide both theoretical and experimental backing. In particular, we incorporate the Gurobi MIP solver, Blossom V algorithm, and Dijkstra's algorithm into architectures that extract suitable features from raw inputs for the traveling salesman problem, the min-cost perfect matching problem and the shortest path problem. The code is available at https://github.com/martius-lab/blackbox-backprop.

Adam Kukučka, Ondřej Fabián, Vít Musil et al.

Histologic assessment of ulcerative colitis (UC) activity is an important endpoint in clinical trials and routine care, but manual grading with indices such as the Nancy histological index (NHI) is time-consuming and prone to observer variability. While computational pathology methods can automate scoring, many approaches depend on dense region-level annotations, which are costly to obtain, particularly in heterogeneous, multicenter cohorts. We propose a weakly supervised multiple instance learning (MIL) approach for whole-slide images that learns from case- and slide-level NHI labels, leveraging foundation models. Our method targets clinically relevant endpoints, including neutrophilic activity and derived Nancy-low/high groupings, enabling full five-grade NHI prediction. On a multicenter dataset of H&E-stained colon biopsies from three hospitals (2019-2025), we evaluate multiple foundation model encoders and aggregation strategies. We find that foundation model choice and resolution substantially affect performance, with Virchow2 providing the most consistent gains, and that a simple ensembling rule improves five-grade NHI prediction compared to a hierarchical gating baseline. Overall, our results demonstrate that weakly supervised MIL with modern foundation-model representations can provide robust, interpretable UC histology activity assessment in realistic multicenter settings.

Vojtěch Kůr, Adam Bajger, Adam Kukučka et al.

Convolutional neural networks (CNNs) underpin many modern computer vision systems. With applications ranging from common to critical areas, a need to explain and understand the model and its decisions (XAI) emerged. Prior works suggest that in the top layers of CNNs, the individual channels can be attributed to classifying human-understandable concepts. Concept relevance propagation (CRP) methods can backtrack predictions to these channels and find images that most activate these channels. However, current CRP workflows are largely manual: experts must inspect activation images to name the discovered concepts and must synthesize verbose explanations from relevance maps, limiting the accessibility of the explanations and their scalability. To address these issues, we introduce Large Language model EXplaIns COncept Relevance Propagation (LLEXICORP), a modular pipeline that couples CRP with a multimodal large language model. Our approach automatically assigns descriptive names to concept prototypes and generates natural-language explanations that translate quantitative relevance distributions into intuitive narratives. To ensure faithfulness, we craft prompts that teach the language model the semantics of CRP through examples and enforce a separation between naming and explanation tasks. The resulting text can be tailored to different audiences, offering low-level technical descriptions for experts and high-level summaries for non-technical stakeholders. We qualitatively evaluate our method on various images from ImageNet on a VGG16 model. Our findings suggest that integrating concept-based attribution methods with large language models can significantly lower the barrier to interpreting deep neural networks, paving the way for more transparent AI systems.

Adam Bajger, Jan Obdržálek, Vojtěch Kůr et al.

We present a clustering-based explainability technique for digital pathology models based on convolutional neural networks. Unlike commonly used methods based on saliency maps, such as occlusion, GradCAM, or relevance propagation, which highlight regions that contribute the most to the prediction for a single slide, our method shows the global behaviour of the model under consideration, while also providing more fine-grained information. The result clusters can be visualised not only to understand the model, but also to increase confidence in its operation, leading to faster adoption in clinical practice. We also evaluate the performance of our technique on an existing model for detecting prostate cancer, demonstrating its usefulness.

Anselm Paulus, A. René Geist, Pierre Schumacher et al.

Contact forces pose a major challenge for gradient-based optimization of robot dynamics as they introduce jumps in the system's velocities. Penalty-based simulators, such as MuJoCo, simplify gradient computation by softening the contact forces. However, realistically simulating hard contacts requires very stiff contact settings, which leads to incorrect gradients when using automatic differentiation. On the other hand, using non-stiff settings strongly increases the sim-to-real gap. We analyze the contact computation of penalty-based simulators to identify the causes of gradient errors. Then, we propose DiffMJX, which combines adaptive integration with MuJoCo XLA, to notably improve gradient quality in the presence of hard contacts. Finally, we address a key limitation of contact gradients: they vanish when objects do not touch. To overcome this, we introduce Contacts From Distance (CFD), a mechanism that enables the simulator to generate informative contact gradients even before objects are in contact. To preserve physical realism, we apply CFD only in the backward pass using a straight-through trick, allowing us to compute useful gradients without modifying the forward simulation.

Vojtěch Kůr, Vít Musil, Vojtěch Řehák

Adversarial Patrolling games form a subclass of Security games where a Defender moves between locations, guarding vulnerable targets. The main algorithmic problem is constructing a strategy for the Defender that minimizes the worst damage an Attacker can cause. We focus on the class of finite-memory (also known as regular) Defender's strategies that experimentally outperformed other competing classes. A finite-memory strategy can be seen as a positional strategy on a finite set of states. Each state consists of a pair of a location and a certain integer value--called memory. Existing algorithms improve the transitional probabilities between the states but require that the available memory size itself is assigned at each location manually. Choosing the right memory assignment is a well-known open and hard problem that hinders the usability of finite-memory strategies. We solve this issue by developing a general method that iteratively changes the memory assignment. Our algorithm can be used in connection with \emph{any} black-box strategy optimization tool. We evaluate our method on various experiments and show its robustness by solving instances of various patrolling models.

David Klaška, Antonín Kučera, Vojtěch Kůr et al.

The long-run average payoff per transition (mean payoff) is the main tool for specifying the performance and dependability properties of discrete systems. The problem of constructing a controller (strategy) simultaneously optimizing several mean payoffs has been deeply studied for stochastic and game-theoretic models. One common issue of the constructed controllers is the instability of the mean payoffs, measured by the deviations of the average rewards per transition computed in a finite "window" sliding along a run. Unfortunately, the problem of simultaneously optimizing the mean payoffs under local stability constraints is computationally hard, and the existing works do not provide a practically usable algorithm even for non-stochastic models such as two-player games. In this paper, we design and evaluate the first efficient and scalable solution to this problem applicable to Markov decision processes.

Anselm Paulus, Michal Rolínek, Vít Musil et al.

Bridging logical and algorithmic reasoning with modern machine learning techniques is a fundamental challenge with potentially transformative impact. On the algorithmic side, many NP-hard problems can be expressed as integer programs, in which the constraints play the role of their "combinatorial specification." In this work, we aim to integrate integer programming solvers into neural network architectures as layers capable of learning both the cost terms and the constraints. The resulting end-to-end trainable architectures jointly extract features from raw data and solve a suitable (learned) combinatorial problem with state-of-the-art integer programming solvers. We demonstrate the potential of such layers with an extensive performance analysis on synthetic data and with a demonstration on a competitive computer vision keypoint matching benchmark.