PyTorchKR Papers

최신 논문

총 139,336편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2504.00885공식 코드 · Papers with Code

Spectral Architecture Search for Neural Networks

Architecture design and optimization are challenging problems in the field of artificial neural networks. Working in this context, we here present SPARCS (SPectral ARchiteCture Search), a novel architecture search protocol which exploits the spectral attributes of the inter-layer transfer matrices. SPARCS allows one to explore the space of possible architectures by spanning continuous and differentiable manifolds, thus enabling for gradient-based optimization algorithms to be eventually employed. With reference to simple benchmark models, we show that the newly proposed method yields a self-emerging architecture with a minimal degree of expressivity to handle the task under investigation and with a reduced parameter count as compared to other viable alternatives.

arXiv:2504.00746공식 코드 · Papers with Code

Semi-Data-Driven Model Predictive Control: A Physics-Informed Data-Driven Control Approach

Data-enabled predictive control (DeePC) has emerged as a powerful technique to control complex systems without the need for extensive modeling efforts. However, relying solely on offline collected data trajectories to represent the system dynamics introduces certain drawbacks. Therefore, we present a novel semi-data-driven model predictive control (SD-MPC) framework that combines (limited) model information with DeePC to address a range of these drawbacks, including sensitivity to noisy data, lack of robustness, and a high computational burden. In this work we focus on the performance of DeePC in operating regimes not captured by the offline collected data trajectories and demonstrate how incorporating an underlying parametric model can counteract this issue. SD-MPC exhibits equivalent closed-loop performance as DeePC for deterministic linear time-invariant systems. Simulations demonstrate the general control performance of the proposed SD-MPC for both a linear time-invariant system and a nonlinear system modeled as a linear parameter-varying system. These results provide numerical evidence of the enhanced robustness of SD-MPC over classical DeePC.

arXiv:2504.00673공식 코드 · Papers with Code

In-Context Learning for Zero-Shot Speed Estimation of BLDC motors

Accurate speed estimation in sensorless brushless DC motors is essential for high-performance control and monitoring, yet conventional model-based approaches struggle with system nonlinearities and parameter uncertainties. In this work, we propose an in-context learning framework leveraging transformer-based models to perform zero-shot speed estimation using only electrical measurements. By training the filter offline on simulated motor trajectories, we enable real-time inference on unseen real motors without retraining, eliminating the need for explicit system identification while retaining adaptability to varying operating conditions. Experimental results demonstrate that our method outperforms traditional Kalman filter-based estimators, especially in low-speed regimes that are crucial during motor startup.

arXiv:2504.00634공식 코드 · Papers with Code

CNOT-Optimal Clifford Synthesis as SAT

Clifford circuit optimization is an important step in the quantum compilation pipeline. Major compilers employ heuristic approaches. While they are fast, their results are often suboptimal. Minimization of noisy gates, like 2-qubit CNOT gates, is crucial for practical computing. Exact approaches have been proposed to fill the gap left by heuristic approaches. Among these are SAT based approaches that optimize gate count or depth, but they suffer from scalability issues. Further, they do not guarantee optimality on more important metrics like CNOT count or CNOT depth. A recent work proposed an exhaustive search only on Clifford circuits in a certain normal form to guarantee CNOT count optimality. But an exhaustive approach cannot scale beyond 6 qubits. In this paper, we incorporate search restricted to Clifford normal forms in a SAT encoding to guarantee CNOT count optimality. By allowing parallel plans, we propose a second SAT encoding that optimizes CNOT depth. By taking advantage of flexibility in SAT based approaches, we also handle connectivity restrictions in hardware platforms, and allow for qubit relabeling. We have implemented the above encodings and variations in our open source tool Q-Synth. In experiments, our encodings significantly outperform existing SAT approaches on random Clifford circuits. We consider practical VQE and Feynman benchmarks to compare with TKET and Qiskit compilers. In all-to-all connectivity, we observe reductions up to 32.1% in CNOT count and 48.1% in CNOT depth. Overall, we observe better results than TKET in the CNOT count and depth. We also experiment with connectivity restrictions of major quantum platforms. Compared to Qiskit, we observe up to 30.3% CNOT count and 35.9% CNOT depth further reduction.

arXiv:2504.00626공식 코드 · Papers with Code

Probabilistically safe and efficient model-based Reinforcement Learning

This paper proposes tackling safety-critical stochastic Reinforcement Learning (RL) tasks with a samplebased, model-based approach. At the core of the method lies a Model Predictive Control (MPC) scheme that acts as function approximation, providing a model-based predictive control policy. To ensure safety, a probabilistic Control Barrier Function (CBF) is integrated into the MPC controller. A sample-based approach with guarantees is employed to approximate the effects of stochasticies in the optimal control formulation and to guarantee the probabilistic CBF condition. A learnable terminal cost formulation is included in the MPC objective to counterbalance the additional computational burden due to sampling. An RL algorithm is deployed to learn both the terminal cost and the CBF constraint. Results from our numerical experiment on a constrained LTI problem corroborate the effectiveness of the proposed methodology in reducing computation time while preserving control performance and safety.

arXiv:2504.00525공식 코드 · Papers with Code

Robust LiDAR-Camera Calibration with 2D Gaussian Splatting

LiDAR-camera systems have become increasingly popular in robotics recently. A critical and initial step in integrating the LiDAR and camera data is the calibration of the LiDAR-camera system. Most existing calibration methods rely on auxiliary target objects, which often involve complex manual operations, whereas targetless methods have yet to achieve practical effectiveness. Recognizing that 2D Gaussian Splatting (2DGS) can reconstruct geometric information from camera image sequences, we propose a calibration method that estimates LiDAR-camera extrinsic parameters using geometric constraints. The proposed method begins by reconstructing colorless 2DGS using LiDAR point clouds. Subsequently, we update the colors of the Gaussian splats by minimizing the photometric loss. The extrinsic parameters are optimized during this process. Additionally, we address the limitations of the photometric loss by incorporating the reprojection and triangulation losses, thereby enhancing the calibration robustness and accuracy.

arXiv:2504.00513공식 코드 · Papers with Code

Leveraging LLMs for User Stories in AI Systems: UStAI Dataset

AI systems are gaining widespread adoption across various sectors and domains. Creating high-quality AI system requirements is crucial for aligning the AI system with business goals and consumer values and for social responsibility. However, with the uncertain nature of AI systems and the heavy reliance on sensitive data, more research is needed to address the elicitation and analysis of AI systems requirements. With the proprietary nature of many AI systems, there is a lack of open-source requirements artifacts and technical requirements documents for AI systems, limiting broader research and investigation. With Large Language Models (LLMs) emerging as a promising alternative to human-generated text, this paper investigates the potential use of LLMs to generate user stories for AI systems based on abstracts from scholarly papers. We conducted an empirical evaluation using three LLMs and generated $1260$ user stories from $42$ abstracts from $26$ domains. We assess their quality using the Quality User Story (QUS) framework. Moreover, we identify relevant non-functional requirements (NFRs) and ethical principles. Our analysis demonstrates that the investigated LLMs can generate user stories inspired by the needs of various stakeholders, offering a promising approach for generating user stories for research purposes and for aiding in the early requirements elicitation phase of AI systems. We have compiled and curated a collection of stories generated by various LLMs into a dataset (UStAI), which is now publicly available for use.

arXiv:2504.00276공식 코드 · Papers with Code

On-the-fly Surrogation for Complex Nonlinear Dynamics

High-fidelity models are essential for accurately capturing nonlinear system dynamics. However, simulation of these models is often computationally too expensive and, due to their complexity, they are not directly suitable for analysis, control design or real-time applications. Surrogate modelling techniques seek to construct simplified representations of these systems with minimal complexity, but adequate information on the dynamics given a simulation, analysis or synthesis objective at hand. Despite the widespread availability of system linearizations and the growing computational potential of autograd methods, there is no established approach that systematically exploits them to capture the underlying global nonlinear dynamics. This work proposes a novel surrogate modelling approach that can efficiently build a global representation of the dynamics on-the-fly from local system linearizations without ever explicitly computing a model. Using radial basis function interpolation and the second fundamental theorem of calculus, the surrogate model is only computed at its evaluation, enabling rapid computation for simulation and analysis and seamless incorporation of new linearization data. The efficiency and modelling capabilities of the method are demonstrated on simulation examples.

arXiv:2504.00225공식 코드 · Papers with Code

Distributed Model Predictive Control for Dynamic Cooperation of Multi-Agent Systems

We propose a distributed model predictive control (MPC) framework for coordinating heterogeneous, nonlinear multi-agent systems under individual and coupling constraints. The cooperative task is encoded as a shared objective function minimized collectively by the agents. Each agent optimizes an artificial reference as an intermediate step towards the cooperative objective, along with a control input to track it. We establish recursive feasibility, asymptotic stability, and transient performance bounds under suitable assumptions. The solution to the cooperative task is not predetermined but emerges from the optimized interactions of the agents. We demonstrate the framework on numerical examples inspired by satellite constellation control, collision-free narrow-passage traversal, and coordinated quadrotor flight.

arXiv:2504.00196공식 코드 · Papers with Code

Output-feedback model predictive control under dynamic uncertainties using integral quadratic constraints

In this work, we propose an output-feedback tube-based model predictive control (MPC) scheme for linear systems under dynamic uncertainties that are described via integral quadratic constraints (IQC). By leveraging IQCs, a large class of nonlinear and dynamic uncertainties can be addressed. We leverage recent IQC synthesis tools to design a dynamic controller and an observer that are robust to these uncertainties and minimize the size of the resulting constraint tightening in the MPC. Thereby, we show that the robust estimation problem using IQCs with peak-to-peak performance can be convexified. We guarantee recursive feasibility, robust constraint satisfaction, and input-to-state stability of the resulting MPC scheme.

arXiv:2504.00128공식 코드 · Papers with Code

Improving Predictions of Convective Storm Wind Gusts through Statistical Post-Processing of Neural Weather Models

Issuing timely severe weather warnings helps mitigate potentially disastrous consequences. Recent advancements in Neural Weather Models (NWMs) offer a computationally inexpensive and fast approach for forecasting atmospheric environments on a 0.25{\deg} global grid. For thunderstorms, these environments can be empirically post-processed to predict wind gust distributions at specific locations. With the Pangu-Weather NWM, we apply a hierarchy of statistical and deep learning post-processing methods to forecast hourly wind gusts up to three days ahead. To ensure statistical robustness, we constrain our probabilistic forecasts using generalised extreme-value distributions across five regions in Switzerland. Using a convolutional neural network to post-process the predicted atmospheric environment's spatial patterns yields the best results, outperforming direct forecasting approaches across lead times and wind gust speeds. Our results confirm the added value of NWMs for extreme wind forecasting, especially for designing more responsive early-warning systems.

arXiv:2504.00040공식 코드 · Papers with Code

Quantum Methods for Managing Ambiguity in Natural Language Processing

The Categorical Compositional Distributional (DisCoCat) framework models meaning in natural language using the mathematical framework of quantum theory, expressed as formal diagrams. DisCoCat diagrams can be associated with tensor networks and quantum circuits. DisCoCat diagrams have been connected to density matrices in various contexts in Quantum Natural Language Processing (QNLP). Previous use of density matrices in QNLP entails modelling ambiguous words as probability distributions over more basic words (the word \texttt{queen}, e.g., might mean the reigning queen or the chess piece). In this article, we investigate using probability distributions over processes to account for syntactic ambiguity in sentences. The meanings of these sentences are represented by density matrices. We show how to create probability distributions on quantum circuits that represent the meanings of sentences and explain how this approach generalises tasks from the literature. We conduct an experiment to validate the proposed theory.

arXiv:2504.01915공식 코드 · Papers with Code

Overcoming Deceptiveness in Fitness Optimization with Unsupervised Quality-Diversity

Policy optimization seeks the best solution to a control problem according to an objective or fitness function, serving as a fundamental field of engineering and research with applications in robotics. Traditional optimization methods like reinforcement learning and evolutionary algorithms struggle with deceptive fitness landscapes, where following immediate improvements leads to suboptimal solutions. Quality-diversity (QD) algorithms offer a promising approach by maintaining diverse intermediate solutions as stepping stones for escaping local optima. However, QD algorithms require domain expertise to define hand-crafted features, limiting their applicability where characterizing solution diversity remains unclear. In this paper, we show that unsupervised QD algorithms - specifically the AURORA framework, which learns features from sensory data - efficiently solve deceptive optimization problems without domain expertise. By enhancing AURORA with contrastive learning and periodic extinction events, we propose AURORA-XCon, which outperforms all traditional optimization baselines and matches, in some cases even improving by up to 34%, the best QD baseline with domain-specific hand-crafted features. This work establishes a novel application of unsupervised QD algorithms, shifting their focus from discovering novel solutions toward traditional optimization and expanding their potential to domains where defining feature spaces poses challenges.

arXiv:2504.01786공식 코드 · Papers with Code

BlenderGym: Benchmarking Foundational Model Systems for Graphics Editing

3D graphics editing is crucial in applications like movie production and game design, yet it remains a time-consuming process that demands highly specialized domain expertise. Automating this process is challenging because graphical editing requires performing a variety of tasks, each requiring distinct skill sets. Recently, vision-language models (VLMs) have emerged as a powerful framework for automating the editing process, but their development and evaluation are bottlenecked by the lack of a comprehensive benchmark that requires human-level perception and presents real-world editing complexity. In this work, we present BlenderGym, the first comprehensive VLM system benchmark for 3D graphics editing. BlenderGym evaluates VLM systems through code-based 3D reconstruction tasks. We evaluate closed- and open-source VLM systems and observe that even the state-of-the-art VLM system struggles with tasks relatively easy for human Blender users. Enabled by BlenderGym, we study how inference scaling techniques impact VLM's performance on graphics editing tasks. Notably, our findings reveal that the verifier used to guide the scaling of generation can itself be improved through inference scaling, complementing recent insights on inference scaling of LLM generation in coding and math tasks. We further show that inference compute is not uniformly effective and can be optimized by strategically distributing it between generation and verification.

arXiv:2504.01655공식 코드 · Papers with Code

Q-Adapt: Adapting LMM for Visual Quality Assessment with Progressive Instruction Tuning

The rapid advancement of Large Multi-modal Foundation Models (LMM) has paved the way for the possible Explainable Image Quality Assessment (EIQA) with instruction tuning from two perspectives: overall quality explanation, and attribute-wise perception answering. However, existing works usually overlooked the conflicts between these two types of perception explanations during joint instruction tuning, leading to insufficient perception understanding. To mitigate this, we propose a new paradigm for perception-oriented instruction tuning, i.e., Q-Adapt, which aims to eliminate the conflicts and achieve the synergy between these two EIQA tasks when adapting LMM, resulting in enhanced multi-faceted explanations of IQA. Particularly, we propose a progressive instruction tuning strategy by dividing the adaption process of LMM for EIQA into two stages, where the first stage empowers the LMM with universal perception knowledge tailored for two tasks using an efficient transfer learning strategy, i.e., LoRA, and the second stage introduces the instruction-adaptive visual prompt tuning to dynamically adapt visual features for the different instructions from two tasks. In this way, our proposed Q-Adapt can achieve a lightweight visual quality evaluator, demonstrating comparable performance and, in some instances, superior results across perceptual-related benchmarks and commonly-used IQA databases. The source code is publicly available at https://github.com/yeppp27/Q-Adapt.

arXiv:2504.01408공식 코드 · Papers with Code

From Shadows to Safety: Occlusion Tracking and Risk Mitigation for Urban Autonomous Driving

Autonomous vehicles (AVs) must navigate dynamic urban environments where occlusions and perception limitations introduce significant uncertainties. This research builds upon and extends existing approaches in risk-aware motion planning and occlusion tracking to address these challenges. While prior studies have developed individual methods for occlusion tracking and risk assessment, a comprehensive method integrating these techniques has not been fully explored. We, therefore, enhance a phantom agent-centric model by incorporating sequential reasoning to track occluded areas and predict potential hazards. Our model enables realistic scenario representation and context-aware risk evaluation by modeling diverse phantom agents, each with distinct behavior profiles. Simulations demonstrate that the proposed approach improves situational awareness and balances proactive safety with efficient traffic flow. While these results underline the potential of our method, validation in real-world scenarios is necessary to confirm its feasibility and generalizability. By utilizing and advancing established methodologies, this work contributes to safer and more reliable AV planning in complex urban environments. To support further research, our method is available as open-source software at: https://github.com/TUM-AVS/OcclusionAwareMotionPlanning

arXiv:2504.01261공식 코드 · Papers with Code

ForestVO: Enhancing Visual Odometry in Forest Environments through ForestGlue

Recent advancements in visual odometry systems have improved autonomous navigation; however, challenges persist in complex environments like forests, where dense foliage, variable lighting, and repetitive textures compromise feature correspondence accuracy. To address these challenges, we introduce ForestGlue, enhancing the SuperPoint feature detector through four configurations - grayscale, RGB, RGB-D, and stereo-vision - optimised for various sensing modalities. For feature matching, we employ LightGlue or SuperGlue, retrained with synthetic forest data. ForestGlue achieves comparable pose estimation accuracy to baseline models but requires only 512 keypoints - just 25% of the baseline's 2048 - to reach an LO-RANSAC AUC score of 0.745 at a 10{\deg} threshold. With only a quarter of keypoints needed, ForestGlue significantly reduces computational overhead, demonstrating effectiveness in dynamic forest environments, and making it suitable for real-time deployment on resource-constrained platforms. By combining ForestGlue with a transformer-based pose estimation model, we propose ForestVO, which estimates relative camera poses using matched 2D pixel coordinates between frames. On challenging TartanAir forest sequences, ForestVO achieves an average relative pose error (RPE) of 1.09 m and a kitti_score of 2.33%, outperforming direct-based methods like DSO by 40% in dynamic scenes. Despite using only 10% of the dataset for training, ForestVO maintains competitive performance with TartanVO while being a significantly lighter model. This work establishes an end-to-end deep learning pipeline specifically tailored for visual odometry in forested environments, leveraging forest-specific training data to optimise feature correspondence and pose estimation, thereby enhancing the accuracy and robustness of autonomous navigation systems.

arXiv:2504.01212공식 코드 · Papers with Code

Cooper: A Library for Constrained Optimization in Deep Learning

Cooper is an open-source package for solving constrained optimization problems involving deep learning models. Cooper implements several Lagrangian-based first-order update schemes, making it easy to combine constrained optimization algorithms with high-level features of PyTorch such as automatic differentiation, and specialized deep learning architectures and optimizers. Although Cooper is specifically designed for deep learning applications where gradients are estimated based on mini-batches, it is suitable for general non-convex continuous constrained optimization. Cooper's source code is available at https://github.com/cooper-org/cooper.

arXiv:2504.01201공식 코드 · Papers with Code

Medical large language models are easily distracted

Large language models (LLMs) have the potential to transform medicine, but real-world clinical scenarios contain extraneous information that can hinder performance. The rise of assistive technologies like ambient dictation, which automatically generates draft notes from live patient encounters, has the potential to introduce additional noise making it crucial to assess the ability of LLM's to filter relevant data. To investigate this, we developed MedDistractQA, a benchmark using USMLE-style questions embedded with simulated real-world distractions. Our findings show that distracting statements (polysemous words with clinical meanings used in a non-clinical context or references to unrelated health conditions) can reduce LLM accuracy by up to 17.9%. Commonly proposed solutions to improve model performance such as retrieval-augmented generation (RAG) and medical fine-tuning did not change this effect and in some cases introduced their own confounders and further degraded performance. Our findings suggest that LLMs natively lack the logical mechanisms necessary to distinguish relevant from irrelevant clinical information, posing challenges for real-world applications. MedDistractQA and our results highlights the need for robust mitigation strategies to enhance LLM resilience to extraneous information.

arXiv:2504.01174공식 코드 · Papers with Code

Value Iteration for Learning Concurrently Executable Robotic Control Tasks

Many modern robotic systems such as multi-robot systems and manipulators exhibit redundancy, a property owing to which they are capable of executing multiple tasks. This work proposes a novel method, based on the Reinforcement Learning (RL) paradigm, to train redundant robots to be able to execute multiple tasks concurrently. Our approach differs from typical multi-objective RL methods insofar as the learned tasks can be combined and executed in possibly time-varying prioritized stacks. We do so by first defining a notion of task independence between learned value functions. We then use our definition of task independence to propose a cost functional that encourages a policy, based on an approximated value function, to accomplish its control objective while minimally interfering with the execution of higher priority tasks. This allows us to train a set of control policies that can be executed simultaneously. We also introduce a version of fitted value iteration to learn to approximate our proposed cost functional efficiently. We demonstrate our approach on several scenarios and robotic systems.