PyTorchKR Papers

최신 논문

총 3,273편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2504.15021

Is Intelligence the Right Direction in New OS Scheduling for Multiple Resources in Cloud Environments?

Making it intelligent is a promising way in System/OS design. This paper proposes OSML+, a new ML-based resource scheduling mechanism for co-located cloud services. OSML+ intelligently schedules the cache and main memory bandwidth resources at the memory hierarchy and the computing core resources simultaneously. OSML+ uses a multi-model collaborative learning approach during its scheduling and thus can handle complicated cases, e.g., avoiding resource cliffs, sharing resources among applications, enabling different scheduling policies for applications with different priorities, etc. OSML+ can converge faster using ML models than previous studies. Moreover, OSML+ can automatically learn on the fly and handle dynamically changing workloads accordingly. Using transfer learning technologies, we show our design can work well across various cloud servers, including the latest off-the-shelf large-scale servers. Our experimental results show that OSML+ supports higher loads and meets QoS targets with lower overheads than previous studies.

arXiv:2504.14946

Symmetry-Preserving Architecture for Multi-NUMA Environments (SPANE): A Deep Reinforcement Learning Approach for Dynamic VM Scheduling

As cloud computing continues to evolve, the adoption of multi-NUMA (Non-Uniform Memory Access) architecture by cloud service providers has introduced new challenges in virtual machine (VM) scheduling. To address these challenges and more accurately reflect the complexities faced by modern cloud environments, we introduce the Dynamic VM Allocation problem in Multi-NUMA PM (DVAMP). We formally define both offline and online versions of DVAMP as mixed-integer linear programming problems, providing a rigorous mathematical foundation for analysis. A tight performance bound for greedy online algorithms is derived, offering insights into the worst-case optimality gap as a function of the number of physical machines and VM lifetime variability. To address the challenges posed by DVAMP, we propose SPANE (Symmetry-Preserving Architecture for Multi-NUMA Environments), a novel deep reinforcement learning approach that exploits the problem's inherent symmetries. SPANE produces invariant results under arbitrary permutations of physical machine states, enhancing learning efficiency and solution quality. Extensive experiments conducted on the Huawei-East-1 dataset demonstrate that SPANE outperforms existing baselines, reducing average VM wait time by 45%. Our work contributes to the field of cloud resource management by providing both theoretical insights and practical solutions for VM scheduling in multi-NUMA environments, addressing a critical gap in the literature and offering improved performance for real-world cloud systems.

arXiv:2504.09046

An Enhanced Iterative Deepening Search Algorithm for the Unrestricted Container Rehandling Problem

In container terminal yards, the Container Rehandling Problem (CRP) involves rearranging containers between stacks under specific operational rules, and it is a pivotal optimization challenge in intelligent container scheduling systems. Existing CRP studies primarily focus on minimizing reallocation costs using two-dimensional bay structures, considering factors such as container size, weight, arrival sequences, and retrieval priorities. This paper introduces an enhanced deepening search algorithm integrated with improved lower bounds to boost search efficiency. To further reduce the search space, we design mutually consistent pruning rules to avoid excessive computational overhead. The proposed algorithm is validated on three widely used benchmark datasets for the Unrestricted Container Rehandling Problem (UCRP). Experimental results demonstrate that our approach outperforms state-of-the-art exact algorithms in solving the more general UCRP variant, particularly exhibiting superior efficiency when handling containers within the same priority group under strict time constraints.

arXiv:2504.14773

PLANET: A Collection of Benchmarks for Evaluating LLMs' Planning Capabilities

Planning is central to agents and agentic AI. The ability to plan, e.g., creating travel itineraries within a budget, holds immense potential in both scientific and commercial contexts. Moreover, optimal plans tend to require fewer resources compared to ad-hoc methods. To date, a comprehensive understanding of existing planning benchmarks appears to be lacking. Without it, comparing planning algorithms' performance across domains or selecting suitable algorithms for new scenarios remains challenging. In this paper, we examine a range of planning benchmarks to identify commonly used testbeds for algorithm development and highlight potential gaps. These benchmarks are categorized into embodied environments, web navigation, scheduling, games and puzzles, and everyday task automation. Our study recommends the most appropriate benchmarks for various algorithms and offers insights to guide future benchmark development.

arXiv:2504.14145

PipeWeaver: Addressing Data Dynamicity in Large Multimodal Model Training with Dynamic Interleaved Pipeline

Large multimodal models (LMMs) have demonstrated excellent capabilities in both understanding and generation tasks with various modalities. While these models can accept flexible combinations of input data, their training efficiency suffers from two major issues: pipeline stage imbalance caused by heterogeneous model architectures, and training data dynamicity stemming from the diversity of multimodal data. In this paper, we present PipeWeaver, a dynamic pipeline scheduling framework designed for LMM training. The core of PipeWeaver is dynamic interleaved pipeline, which searches for pipeline schedules dynamically tailored to current training batches. PipeWeaver addresses issues of LMM training with two techniques: adaptive modality-aware partitioning and efficient pipeline schedule search within a hierarchical schedule space. Meanwhile, PipeWeaver utilizes SEMU (Step Emulator), a training simulator for multimodal models, for accurate performance estimations, accelerated by spatial-temporal subgraph reuse to improve search efficiency. Experiments show that PipeWeaver can enhance LMM training efficiency by up to 97.3% compared to state-of-the-art systems, and demonstrate excellent adaptivity to LMM training's data dynamicity.

arXiv:2504.13624

PV-VLM: A Multimodal Vision-Language Approach Incorporating Sky Images for Intra-Hour Photovoltaic Power Forecasting

The rapid proliferation of solar energy has significantly expedited the integration of photovoltaic (PV) systems into contemporary power grids. Considering that the cloud dynamics frequently induce rapid fluctuations in solar irradiance, accurate intra-hour forecasting is critical for ensuring grid stability and facilitating effective energy management. To leverage complementary temporal, textual, and visual information, this paper has proposed PV-VLM, a multimodal forecasting framework that integrates temporal, textual, and visual information by three modules. The Time-Aware Module employed a PatchTST-inspired Transformer to capture both local and global dependencies in PV power time series. Meanwhile, the Prompt-Aware Module encodes textual prompts from historical statistics and dataset descriptors via a large language model. Additionally, the Vision-Aware Module utilizes a pretrained vision-language model to extract high-level semantic features from sky images, emphasizing cloud motion and irradiance fluctuations. The proposed PV-VLM is evaluated using data from a 30-kW rooftop array at Stanford University and through a transfer study on PV systems at the University of Wollongong in Australia. Comparative experiments reveal an average RMSE reduction of approximately 5% and a MAE improvement of nearly 6%, while the transfer study shows average RMSE and MAE reductions of about 7% and 9.5%, respectively. Overall, PV-VLM leverages complementary modalities to provide a robust solution for grid scheduling and energy market participation, enhancing the stability and reliability of PV integration.

arXiv:2504.12156

Predictive Multiplicity in Survival Models: A Method for Quantifying Model Uncertainty in Predictive Maintenance Applications

In many applications, especially those involving prediction, models may yield near-optimal performance yet significantly disagree on individual-level outcomes. This phenomenon, known as predictive multiplicity, has been formally defined in binary, probabilistic, and multi-target classification, and undermines the reliability of predictive systems. However, its implications remain unexplored in the context of survival analysis, which involves estimating the time until a failure or similar event while properly handling censored data. We frame predictive multiplicity as a critical concern in survival-based models and introduce formal measures -- ambiguity, discrepancy, and obscurity -- to quantify it. This is particularly relevant for downstream tasks such as maintenance scheduling, where precise individual risk estimates are essential. Understanding and reporting predictive multiplicity helps build trust in models deployed in high-stakes environments. We apply our methodology to benchmark datasets from predictive maintenance, extending the notion of multiplicity to survival models. Our findings show that ambiguity steadily increases, reaching up to 40-45% of observations; discrepancy is lower but exhibits a similar trend; and obscurity remains mild and concentrated in a few models. These results demonstrate that multiple accurate survival models may yield conflicting estimations of failure risk and degradation progression for the same equipment. This highlights the need to explicitly measure and communicate predictive multiplicity to ensure reliable decision-making in process health management.

arXiv:2504.13236

NNTile: a machine learning framework capable of training extremely large GPT language models on a single node

This study presents an NNTile framework for training large deep neural networks in heterogeneous clusters. The NNTile is based on a StarPU library, which implements task-based parallelism and schedules all provided tasks onto all available processing units (CPUs and GPUs). It means that a particular operation, necessary to train a large neural network, can be performed on any of the CPU cores or GPU devices, depending on automatic scheduling decisions. Such an approach shifts the burden of deciding where to compute and when to communicate from a human being to an automatic decision maker, whether a simple greedy heuristic or a complex AI-based software. The performance of the presented tool for training large language models is demonstrated in extensive numerical experiments.

arXiv:2504.13612

Entropic Time Schedulers for Generative Diffusion Models

The practical performance of generative diffusion models depends on the appropriate choice of the noise scheduling function, which can also be equivalently expressed as a time reparameterization. In this paper, we present a time scheduler that selects sampling points based on entropy rather than uniform time spacing, ensuring that each point contributes an equal amount of information to the final generation. We prove that this time reparameterization does not depend on the initial choice of time. Furthermore, we provide a tractable exact formula to estimate this \emph{entropic time} for a trained model using the training loss without substantial overhead. Alongside the entropic time, inspired by the optimality results, we introduce a rescaled entropic time. In our experiments with mixtures of Gaussian distributions and ImageNet, we show that using the (rescaled) entropic times greatly improves the inference performance of trained models. In particular, we found that the image quality in pretrained EDM2 models, as evaluated by FID and FD-DINO scores, can be substantially increased by the rescaled entropic time reparameterization without increasing the number of function evaluations, with greater improvements in the few NFEs regime.

arXiv:2504.06319

Accelerating LLM Inference Throughput via Asynchronous KV Cache Prefetching

Large Language Models (LLMs) exhibit pronounced memory-bound characteristics during inference due to High Bandwidth Memory (HBM) bandwidth constraints. In this paper, we propose an L2 Cache-oriented asynchronous KV Cache prefetching method to break through the memory bandwidth bottleneck in LLM inference through computation-load overlap. By strategically scheduling idle memory bandwidth during active computation windows, our method proactively prefetches required KV Cache into GPU L2 cache, enabling high-speed L2 cache hits for subsequent accesses and effectively hiding HBM access latency within computational cycles. Extensive experiments on NVIDIA H20 GPUs demonstrate that the proposed method achieves 2.15x improvement in attention kernel efficiency and up to 1.97x end-to-end throughput enhancement, surpassing state-of-the-art baseline FlashAttention-3. Notably, our solution maintains orthogonality to existing optimization techniques and can be integrated with current inference frameworks, providing a scalable latency-hiding solution for next-generation LLM inference engines.

arXiv:2504.12181

Battery-aware Cyclic Scheduling in Energy-harvesting Federated Learning

Federated Learning (FL) has emerged as a promising framework for distributed learning, but its growing complexity has led to significant energy consumption, particularly from computations on the client side. This challenge is especially critical in energy-harvesting FL (EHFL) systems, where device availability fluctuates due to limited and time-varying energy resources. We propose FedBacys, a battery-aware FL framework that introduces cyclic client participation based on users' battery levels to cope with these issues. FedBacys enables clients to save energy and strategically perform local training just before their designated transmission time by clustering clients and scheduling their involvement sequentially. This design minimizes redundant computation, reduces system-wide energy usage, and improves learning stability. Our experiments demonstrate that FedBacys outperforms existing approaches in terms of energy efficiency and performance consistency, exhibiting robustness even under non-i.i.d. training data distributions and with very infrequent battery charging. This work presents the first comprehensive evaluation of cyclic client participation in EHFL, incorporating both communication and computation costs into a unified, resource-aware scheduling strategy.

arXiv:2504.11280

PGU-SGP: A Pheno-Geno Unified Surrogate Genetic Programming For Real-life Container Terminal Truck Scheduling

Data-driven genetic programming (GP) has proven highly effective in solving combinatorial optimization problems under dynamic and uncertain environments. A central challenge lies in fast fitness evaluations on large training datasets, especially for complex real-world problems involving time-consuming simulations. Surrogate models, like phenotypic characterization (PC)-based K-nearest neighbors (KNN), have been applied to reduce computational cost. However, the PC-based similarity measure is confined to behavioral characteristics, overlooking genotypic differences, which can limit surrogate quality and impair performance. To address these issues, this paper proposes a pheno-geno unified surrogate GP algorithm, PGU-SGP, integrating phenotypic and genotypic characterization (GC) to enhance surrogate sample selection and fitness prediction. A novel unified similarity metric combining PC and GC distances is proposed, along with an effective and efficient GC representation. Experimental results of a real-life vehicle scheduling problem demonstrate that PGU-SGP reduces training time by approximately 76% while achieving comparable performance to traditional GP. With the same training time, PGU-SGP significantly outperforms traditional GP and the state-of-the-art algorithm on most datasets. Additionally, PGU-SGP shows faster convergence and improved surrogate quality by maintaining accurate fitness rankings and appropriate selection pressure, further validating its effectiveness.

arXiv:2504.11440공식 코드 · Papers with Code

Greedy Restart Schedules: A Baseline for Dynamic Algorithm Selection on Numerical Black-box Optimization Problems

In many optimization domains, there are multiple different solvers that contribute to the overall state-of-the-art, each performing better on some, and worse on other types of problem instances. Meta-algorithmic approaches, such as instance-based algorithm selection, configuration and scheduling, aim to close this gap by extracting the most performance possible from a set of (configurable) optimizers. In this context, the best performing individual algorithms are often hand-crafted hybrid heuristics which perform many restarts of fast local optimization approaches. However, data-driven techniques to create optimized restart schedules have not yet been extensively studied. Here, we present a simple scheduling approach that iteratively selects the algorithm performing best on the distribution of unsolved training problems at time of selection, resulting in a problem-independent solver schedule. We demonstrate our approach using well-known optimizers from numerical black-box optimization on the BBOB testbed, bridging much of the gap between single and virtual best solver from the original portfolio across various evaluation protocols. Our greedy restart schedule presents a powerful baseline for more complex dynamic algorithm selection models.

arXiv:2504.10648공식 코드 · Papers with Code

Modeling and solving an integrated periodic vehicle routing and capacitated facility location problem in the context of solid waste collection

Few activities are as crucial in urban environments as waste management. Mismanagement of waste can cause significant economic, social, and environmental damage. However, waste management is often a complex system to manage and therefore where computational decision-support tools can play a pivotal role in assisting managers to make faster and better decisions. In this sense, this article proposes, on the one hand, a unified optimization model to address two common waste management system optimization problem: the determination of the capacity of waste bins in the collection network and the design and scheduling of collection routes. The integration of these two problems is not usual in the literature since each of them separately is already a major computational challenge. On the other hand, two improved exact formulations based on mathematical programming and a genetic algorithm (GA) are provided to solve this proposed unified optimization model. It should be noted that the GA considers a mixed chromosome representation of the solutions combining binary and integer alleles, in order to solve realistic instances of this complex problem. Also, different genetic operators have been tested to study which combination of them obtained better results in execution times on the order of that of the exact solvers. The obtained results show that the proposed GA is able to match the results of exact solvers on small instances and, in addition, can obtain feasible solutions on large instances, where exact formulations are not applicable, in reasonable computation times.

arXiv:2504.09680

SPOT: Spatio-Temporal Pattern Mining and Optimization for Load Consolidation in Freight Transportation Networks

Freight consolidation has significant potential to reduce transportation costs and mitigate congestion and pollution. An effective load consolidation plan relies on carefully chosen consolidation points to ensure alignment with existing transportation management processes, such as driver scheduling, personnel planning, and terminal operations. This complexity represents a significant challenge when searching for optimal consolidation strategies. Traditional optimization-based methods provide exact solutions, but their computational complexity makes them impractical for large-scale instances and they fail to leverage historical data. Machine learning-based approaches address these issues but often ignore operational constraints, leading to infeasible consolidation plans. This work proposes SPOT, an end-to-end approach that integrates the benefits of machine learning (ML) and optimization for load consolidation. The ML component plays a key role in the planning phase by identifying the consolidation points through spatio-temporal clustering and constrained frequent itemset mining, while the optimization selects the most cost-effective feasible consolidation routes for a given operational day. Extensive experiments conducted on industrial load data demonstrate that SPOT significantly reduces travel distance and transportation costs (by about 50% on large terminals) compared to the existing industry-standard load planning strategy and a neighborhood-based heuristic. Moreover, the ML component provides valuable tactical-level insights by identifying frequently recurring consolidation opportunities that guide proactive planning. In addition, SPOT is computationally efficient and can be easily scaled to accommodate large transportation networks.

arXiv:2504.09035공식 코드 · Papers with Code

InterQ: A DQN Framework for Optimal Intermittent Control

In this letter, we explore the communication-control co-design of discrete-time stochastic linear systems through reinforcement learning. Specifically, we examine a closed-loop system involving two sequential decision-makers: a scheduler and a controller. The scheduler continuously monitors the system's state but transmits it to the controller intermittently to balance the communication cost and control performance. The controller, in turn, determines the control input based on the intermittently received information. Given the partially nested information structure, we show that the optimal control policy follows a certainty-equivalence form. Subsequently, we analyze the qualitative behavior of the scheduling policy. To develop the optimal scheduling policy, we propose InterQ, a deep reinforcement learning algorithm which uses a deep neural network to approximate the Q-function. Through extensive numerical evaluations, we analyze the scheduling landscape and further compare our approach against two baseline strategies: (a) a multi-period periodic scheduling policy, and (b) an event-triggered policy. The results demonstrate that our proposed method outperforms both baselines. The open source implementation can be found at https://github.com/AC-sh/InterQ.

arXiv:2504.09583

AirVista-II: An Agentic System for Embodied UAVs Toward Dynamic Scene Semantic Understanding

Unmanned Aerial Vehicles (UAVs) are increasingly important in dynamic environments such as logistics transportation and disaster response. However, current tasks often rely on human operators to monitor aerial videos and make operational decisions. This mode of human-machine collaboration suffers from significant limitations in efficiency and adaptability. In this paper, we present AirVista-II -- an end-to-end agentic system for embodied UAVs, designed to enable general-purpose semantic understanding and reasoning in dynamic scenes. The system integrates agent-based task identification and scheduling, multimodal perception mechanisms, and differentiated keyframe extraction strategies tailored for various temporal scenarios, enabling the efficient capture of critical scene information. Experimental results demonstrate that the proposed system achieves high-quality semantic understanding across diverse UAV-based dynamic scenarios under a zero-shot setting.

arXiv:2504.11320공식 코드 · Papers with Code

Optimizing LLM Inference: Fluid-Guided Online Scheduling with Memory Constraints

Large Language Models (LLMs) are indispensable in today's applications, but their inference procedure -- generating responses by processing text in segments and using a memory-heavy Key-Value (KV) cache -- demands significant computational resources, particularly under memory constraints. This paper formulates LLM inference optimization as a multi-stage online scheduling problem where sequential prompt arrivals and KV cache growth render conventional scheduling ineffective. We develop a fluid dynamics approximation to provide a tractable benchmark that guides algorithm design. Building on this, we propose the Waiting for Accumulated Inference Threshold (WAIT) algorithm, which uses multiple thresholds to schedule incoming prompts optimally when output lengths are known, and extend it to Nested WAIT for cases with unknown output lengths. Theoretical analysis shows that both algorithms achieve near-optimal performance against the fluid benchmark in heavy traffic conditions, balancing throughput, latency, and Time to First Token (TTFT). Experiments with the Llama-7B model on an A100 GPU using both synthetic and real-world datasets demonstrate improved throughput and reduced latency relative to established baselines like vLLM and Sarathi. This work bridges operations research and machine learning, offering a rigorous framework for the efficient deployment of LLMs under memory constraints.

arXiv:2504.11197

Efficient Distributed Retrieval-Augmented Generation for Enhancing Language Model Performance

Small language models (SLMs) support efficient deployments on resource-constrained edge devices, but their limited capacity compromises inference performance. Retrieval-augmented generation (RAG) is a promising solution to enhance model performance by integrating external databases, without requiring intensive on-device model retraining. However, large-scale public databases and user-specific private contextual documents are typically located on the cloud and the device separately, while existing RAG implementations are primarily centralized. To bridge this gap, we propose DRAGON, a distributed RAG framework to enhance on-device SLMs through both general and personal knowledge without the risk of leaking document privacy. Specifically, DRAGON decomposes multi-document RAG into multiple parallel token generation processes performed independently and locally on the cloud and the device, and employs a newly designed Speculative Aggregation, a dual-side speculative algorithm to avoid frequent output synchronization between the cloud and device. A new scheduling algorithm is further introduced to identify the optimal aggregation side based on real-time network conditions. Evaluations on real-world hardware testbed demonstrate a significant performance improvement of DRAGON-up to 1.9x greater gains over standalone SLM compared to the centralized RAG, substantial reduction in per-token latency, and negligible Time to First Token (TTFT) overhead.

arXiv:2504.08850

SpecEE: Accelerating Large Language Model Inference with Speculative Early Exiting

Early exiting has recently emerged as a promising technique for accelerating large language models (LLMs) by effectively reducing the hardware computation and memory access. In this paper, we present SpecEE, a fast LLM inference engine with speculative early exiting. (1) At the algorithm level, we propose the speculation-based lightweight predictor design by exploiting the probabilistic correlation between the speculative tokens and the correct results and high parallelism of GPUs. (2) At the system level, we point out that not all layers need a predictor and design the two-level heuristic predictor scheduling engine based on skewed distribution and contextual similarity. (3) At the mapping level, we point out that different decoding methods share the same essential characteristics, and propose the context-aware merged mapping for predictor with efficient GPU implementations to support speculative decoding, and form a framework for various existing orthogonal acceleration techniques (e.g., quantization and sparse activation) on cloud and personal computer (PC) scenarios, successfully pushing the Pareto frontier of accuracy and speedup. It is worth noting that SpecEE can be applied to any LLM by negligible training overhead in advance without affecting the model original parameters. Extensive experiments show that SpecEE achieves 2.25x and 2.43x speedup with Llama2-7B on cloud and PC scenarios respectively.