Wan Du

Xianzhong Ding, Le Chen, Murali Emani et al.

Large Language Models (LLMs), including the LLaMA model, have exhibited their efficacy across various general-domain natural language processing (NLP) tasks. However, their performance in high-performance computing (HPC) domain tasks has been less than optimal due to the specialized expertise required to interpret the model responses. In response to this challenge, we propose HPC-GPT, a novel LLaMA-based model that has been supervised fine-tuning using generated QA (Question-Answer) instances for the HPC domain. To evaluate its effectiveness, we concentrate on two HPC tasks: managing AI models and datasets for HPC, and data race detection. By employing HPC-GPT, we demonstrate comparable performance with existing methods on both tasks, exemplifying its excellence in HPC-related scenarios. Our experiments on open-source benchmarks yield extensive results, underscoring HPC-GPT's potential to bridge the performance gap between LLMs and HPC-specific tasks. With HPC-GPT, we aim to pave the way for LLMs to excel in HPC domains, simplifying the utilization of language models in complex computing applications.

Yuning Chen, Kang Yang, Zhiyu An et al.

The rapid decline in groundwater around the world poses a significant challenge to sustainable agriculture. To address this issue, agricultural managed aquifer recharge (Ag-MAR) is proposed to recharge the aquifer by artificially flooding agricultural lands using surface water. Ag-MAR requires a carefully selected flooding schedule to avoid affecting the oxygen absorption of crop roots. However, current Ag-MAR scheduling does not take into account complex environmental factors such as weather and soil oxygen, resulting in crop damage and insufficient recharging amounts. This paper proposes MARLP, the first end-to-end data-driven control system for Ag-MAR. We first formulate Ag-MAR as an optimization problem. To that end, we analyze four-year in-field datasets, which reveal the multi-periodicity feature of the soil oxygen level trends and the opportunity to use external weather forecasts and flooding proposals as exogenous clues for soil oxygen prediction. Then, we design a two-stage forecasting framework. In the first stage, it extracts both the cross-variate dependency and the periodic patterns from historical data to conduct preliminary forecasting. In the second stage, it uses weather-soil and flooding-soil causality to facilitate an accurate prediction of soil oxygen levels. Finally, we conduct model predictive control (MPC) for Ag-MAR flooding. To address the challenge of large action spaces, we devise a heuristic planning module to reduce the number of flooding proposals to enable the search for optimal solutions. Real-world experiments show that MARLP reduces the oxygen deficit ratio by 86.8% while improving the recharging amount in unit time by 35.8%, compared with the previous four years.

Zhiyu An, Xianzhong Ding, Wan Du

Recent research has shown the potential of Model-based Reinforcement Learning (MBRL) to enhance energy efficiency of Heating, Ventilation, and Air Conditioning (HVAC) systems. However, existing methods rely on black-box thermal dynamics models and stochastic optimizers, lacking reliability guarantees and posing risks to occupant health. In this work, we overcome the reliability bottleneck by redesigning HVAC controllers using decision trees extracted from existing thermal dynamics models and historical data. Our decision tree-based policies are deterministic, verifiable, interpretable, and more energy-efficient than current MBRL methods. First, we introduce a novel verification criterion for RL agents in HVAC control based on domain knowledge. Second, we develop a policy extraction procedure that produces a verifiable decision tree policy. We found that the high dimensionality of the thermal dynamics model input hinders the efficiency of policy extraction. To tackle the dimensionality challenge, we leverage importance sampling conditioned on historical data distributions, significantly improving policy extraction efficiency. Lastly, we present an offline verification algorithm that guarantees the reliability of a control policy. Extensive experiments show that our method saves 68.4% more energy and increases human comfort gain by 14.8% compared to the state-of-the-art method, in addition to an 1127x reduction in computation overhead. Our code and data are available at https://github.com/ryeii/Veri_HVAC

Xianzhong Ding, Yunkai Zhang, Binbin Chen et al.

Modern industry-scale data centers need to manage a large number of virtual machines (VMs). Due to the continual creation and release of VMs, many small resource fragments are scattered across physical machines (PMs). To handle these fragments, data centers periodically reschedule some VMs to alternative PMs, a practice commonly referred to as VM rescheduling. Despite the increasing importance of VM rescheduling as data centers grow in size, the problem remains understudied. We first show that, unlike most combinatorial optimization tasks, the inference time of VM rescheduling algorithms significantly influences their performance, due to dynamic VM state changes during this period. This causes existing methods to scale poorly. Therefore, we develop a reinforcement learning system for VM rescheduling, VM2RL, which incorporates a set of customized techniques, such as a two-stage framework that accommodates diverse constraints and workload conditions, a feature extraction module that captures relational information specific to rescheduling, as well as a risk-seeking evaluation enabling users to optimize the trade-off between latency and accuracy. We conduct extensive experiments with data from an industry-scale data center. Our results show that VM2RL can achieve a performance comparable to the optimal solution but with a running time of seconds. Code and datasets are open-sourced: https://github.com/zhykoties/VMR2L_eurosys, https://drive.google.com/drive/folders/1PfRo1cVwuhH30XhsE2Np3xqJn2GpX5qy.

Zhiyu An, Wan Du

Large language models are increasingly influencing human moral decisions, yet current approaches focus primarily on evaluating rather than actively steering their moral decisions. We formulate this as an out-of-distribution moral alignment problem, where LLM agents must learn to apply consistent moral reasoning frameworks to scenarios beyond their training distribution. We introduce Moral-Reason-QA, a novel dataset extending 680 human-annotated, high-ambiguity moral scenarios with framework-specific reasoning traces across utilitarian, deontological, and virtue ethics, enabling systematic evaluation of moral generalization in realistic decision contexts. Our learning approach employs Group Relative Policy Optimization with composite rewards that simultaneously optimize decision alignment and framework-specific reasoning processes to facilitate learning of the underlying moral frameworks. Experimental results demonstrate successful generalization to unseen moral scenarios, with softmax-normalized alignment scores improving by +0.757 for utilitarian and +0.450 for deontological frameworks when tested on out-of-distribution evaluation sets. The experiments also reveal training challenges and promising directions that inform future research. These findings establish that LLM agents can be systematically trained to internalize and apply specific moral frameworks to novel situations, providing a critical foundation for AI safety as language models become more integrated into human decision-making processes.

Sikai Yang, Gang Yan, Wan Du

Losing track of reading progress during line switching can be frustrating. Eye gaze tracking technology offers a potential solution by highlighting read paragraphs, aiding users in avoiding wrong line switches. However, the gap between gaze tracking accuracy (2-3 cm) and text line spacing (3-5 mm) makes direct application impractical. Existing methods leverage the linear reading pattern but fail during jump reading. This paper presents a reading tracking and highlighting system that supports both linear and jump reading. Based on experimental insights from the gaze nature study of 16 users, two gaze error models are designed to enable both jump reading detection and relocation. The system further leverages the large language model's contextual perception capability in aiding reading tracking. A reading tracking domain-specific line-gaze alignment opportunity is also exploited to enable dynamic and frequent calibration of the gaze results. Controlled experiments demonstrate reliable linear reading tracking, as well as 84% accuracy in tracking jump reading. Furthermore, real field tests with 18 volunteers demonstrated the system's effectiveness in tracking and highlighting read paragraphs, improving reading efficiency, and enhancing user experience.

Sikai Yang, Wan Du

As human eyes serve as conduits of rich information, unveiling emotions, intentions, and even aspects of an individual's health and overall well-being, gaze tracking also enables various human-computer interaction applications, as well as insights in psychological and medical research. However, existing gaze tracking solutions fall short at handling free user movement, and also require laborious user effort in system calibration. We introduce Tri-Cam, a practical deep learning-based gaze tracking system using three affordable RGB webcams. It features a split network structure for efficient training, as well as designated network designs to handle the separated gaze tracking tasks. Tri-Cam is also equipped with an implicit calibration module, which makes use of mouse click opportunities to reduce calibration overhead on the user's end. We evaluate Tri-Cam against Tobii, the state-of-the-art commercial eye tracker, achieving comparable accuracy, while supporting a wider free movement area. In conclusion, Tri-Cam provides a user-friendly, affordable, and robust gaze tracking solution that could practically enable various applications.

Kang Yang, Yuning Chen, Wan Du

We present Generalizable Wireless Radiance Fields (GWRF), a framework for modeling wireless signal propagation at arbitrary 3D transmitter and receiver positions. Unlike previous methods that adapt vanilla Neural Radiance Fields (NeRF) from the optical to the wireless signal domain, requiring extensive per-scene training, GWRF generalizes effectively across scenes. First, a geometry-aware Transformer encoder-based wireless scene representation module incorporates information from geographically proximate transmitters to learn a generalizable wireless radiance field. Second, a neural-driven ray tracing algorithm operates on this field to automatically compute signal reception at the receiver. Experimental results demonstrate that GWRF outperforms existing methods on single scenes and achieves state-of-the-art performance on unseen scenes.

Sikai Yang, Kang Yang, Yuning Chen et al.

This work presents ARD2, a framework that enables real-time through-wall surveillance using two aerial drones and an augmented reality (AR) device. ARD2 consists of two main steps: target direction estimation and contour reconstruction. In the first stage, ARD2 leverages geometric relationships between the drones, the user, and the target to project the target's direction onto the user's AR display. In the second stage, images from the drones are synthesized to reconstruct the target's contour, allowing the user to visualize the target behind walls. Experimental results demonstrate the system's accuracy in both direction estimation and contour reconstruction.

Kunpeng Liu, Pengfei Wang, Dongjie Wang et al.

In this paper, we propose a single-agent Monte Carlo based reinforced feature selection (MCRFS) method, as well as two efficiency improvement strategies, i.e., early stopping (ES) strategy and reward-level interactive (RI) strategy. Feature selection is one of the most important technologies in data prepossessing, aiming to find the optimal feature subset for a given downstream machine learning task. Enormous research has been done to improve its effectiveness and efficiency. Recently, the multi-agent reinforced feature selection (MARFS) has achieved great success in improving the performance of feature selection. However, MARFS suffers from the heavy burden of computational cost, which greatly limits its application in real-world scenarios. In this paper, we propose an efficient reinforcement feature selection method, which uses one agent to traverse the whole feature set, and decides to select or not select each feature one by one. Specifically, we first develop one behavior policy and use it to traverse the feature set and generate training data. And then, we evaluate the target policy based on the training data and improve the target policy by Bellman equation. Besides, we conduct the importance sampling in an incremental way, and propose an early stopping strategy to improve the training efficiency by the removal of skew data. In the early stopping strategy, the behavior policy stops traversing with a probability inversely proportional to the importance sampling weight. In addition, we propose a reward-level interactive strategy to improve the training efficiency via reward-level external advice. Finally, we design extensive experiments on real-world data to demonstrate the superiority of the proposed method.

Zhihao Shen, Wan Du, Xi Zhao et al.

Retrieving similar trajectories from a large trajectory dataset is important for a variety of applications, like transportation planning and mobility analysis. Unlike previous works based on fine-grained GPS trajectories, this paper investigates the feasibility of identifying similar trajectories from cellular data observed by mobile infrastructure, which provide more comprehensive coverage. To handle the large localization errors and low sample rates of cellular data, we develop a holistic system, cellSim, which seamlessly integrates map matching and similar trajectory search. A set of map matching techniques are proposed to transform cell tower sequences into moving trajectories on a road map by considering the unique features of cellular data, like the dynamic density of cell towers and bidirectional roads. To further improve the accuracy of similarity search, map matching outputs M trajectory candidates of different confidence, and a new similarity measure scheme is developed to process the map matching results. Meanwhile, M is dynamically adapted to maintain a low false positive rate of the similarity search, and two pruning schemes are proposed to minimize the computation overhead. Extensive experiments on a large-scale dataset and real-world trajectories of 1701 km reveal that cellSim provides high accuracy (precision 62.4% and recall of 89.8%).