arXiv subjects

Yu Liu

Publications and source records attributed to Yu Liu.

At least 631 records · Page 35Linked to original sources

arXiv2024

CSCNET: Class-Specified Cascaded Network for Compositional Zero-Shot Learning

Attribute and object (A-O) disentanglement is a fundamental and critical problem for Compositional Zero-shot Learning (CZSL), whose aim is to recognize novel A-O compositions based on foregone knowledge. Existing methods based on disentangled representation learning lose sight of the contextual dependency between the A-O primitive pairs. Inspired by this, we propose a novel A-O disentangled framework for CZSL, namely Class-specified Cascaded Network (CSCNet). The key insight is to firstly classify one primitive and then specifies the predicted class as a priori for guiding another primitive recognition in a cascaded fashion. To this end, CSCNet constructs Attribute-to-Object and Object-to-Attribute cascaded branches, in addition to a composition branch modeling the two primitives as a whole. Notably, we devise a parametric classifier (ParamCls) to improve the matching between visual and semantic embeddings. By improving the A-O disentanglement, our framework achieves superior results than previous competitive methods.

cs.CV↗

arXiv2024

RAPHAEL: Text-to-Image Generation via Large Mixture of Diffusion Paths

Text-to-image generation has recently witnessed remarkable achievements. We introduce a text-conditional image diffusion model, termed RAPHAEL, to generate highly artistic images, which accurately portray the text prompts, encompassing multiple nouns, adjectives, and verbs. This is achieved by stacking tens of mixture-of-experts (MoEs) layers, i.e., space-MoE and time-MoE layers, enabling billions of diffusion paths (routes) from the network input to the output. Each path intuitively functions as a "painter" for depicting a particular textual concept onto a specified image region at a diffusion timestep. Comprehensive experiments reveal that RAPHAEL outperforms recent cutting-edge models, such as Stable Diffusion, ERNIE-ViLG 2.0, DeepFloyd, and DALL-E 2, in terms of both image quality and aesthetic appeal. Firstly, RAPHAEL exhibits superior performance in switching images across diverse styles, such as Japanese comics, realism, cyberpunk, and ink illustration. Secondly, a single model with three billion parameters, trained on 1,000 A100 GPUs for two months, achieves a state-of-the-art zero-shot FID score of 6.61 on the COCO dataset. Furthermore, RAPHAEL significantly surpasses its counterparts in human evaluation on the ViLG-300 benchmark. We believe that RAPHAEL holds the potential to propel the frontiers of image generation research in both academia and industry, paving the way for future breakthroughs in this rapidly evolving field. More details can be found on a webpage: https://raphael-painter.github.io/.

cs.CV↗

arXiv2024

Enabling 5G QoS configuration capabilities for IoT applications on container orchestration platform

Container orchestration platform is the foundation of modern cloud infrastructure. In recent years, container orchestration platform has been evolving to cross the boundary of device, edge, and cloud. More and more IoT applications such as robotics and XR have been deployed across the device-cloud continuum through the container orchestration platform, e.g., the Kubernetes (K8s) framework. Meanwhile, the rapid expansion of advanced communication technologies like 5G has endorsed the revolution in IoT applications as more network resource is available for critical IoT use cases. This paper aims to integrate network configuration capabilities provided by a 5G Network Exposure Function (NEF) into the K8s framework which is used to simplify application deployment in an orchestration in the device-cloud continuum. Specifically, a Linux fwmark-based network Quality of Service (QoS) configuration method is proposed to expose the QoS information from an overlay network that is used by the container orchestration platform to the underlay network. A Container Network Interface (CNI) plugin-based implementation is demonstrated to perform QoS configuration for the 5G network. The proposed solution is validated with an existing localization and mapping application to verify the feasibility. The proposed solution has the following benefits: (1) The solution is a Kubernetes-native approach which adopts the CNI plugin mechanism. (2) The solution can expose the QoS information from an overlay network to an underlay network in a non-intrusive manner. (3) No packet manipulation is required to greatly reduce the overhead for packet processing. (4) It extends the K8s bandwidth limit feature from on-node to the access network. (5) It is compatible with the 5G infrastructure without any alteration or adding extra complexity.

cs.NI↗

arXiv2024

RoboKube: Establishing a New Foundation for the Cloud Native Evolution in Robotics

Cloud native technologies have been observed to expand into the realm of Internet of Things (IoT) and Cyber-physical Systems, of which an important application domain is robotics. In this paper, we review the cloudification practice in the robotics domain from both literature and industrial perspectives. We propose RoboKube, an adaptive framework that is based on the Kubernetes (K8s) ecosystem to set up a common platform across the device-cloud continuum for the deployment of cloudified Robotic Operating System (ROS) powered applications, to facilitate the cloud native evolution in robotics. We examine the process of modernizing ROS applications using cloud-native technologies, focusing on both the platform and application perspectives. In addition, we address the challenges of networking setups for heterogeneous environments. This paper intends to serves as a guide for developers and researchers, offering insights into containerization strategies, ROS node distribution and clustering, and deployment options. To demonstrate the feasibility of our approach, we present a case study involving the cloudification of a teleoperation testbed.

cs.RO↗

arXiv2024

OTFS vs OFDM: Which is Superior in Multiuser LEO Satellite Communications

Orthogonal time frequency space (OTFS) modulation, a delay-Doppler (DD) domain communication scheme exhibiting strong robustness against the Doppler shifts, has the potentials to be employed in LEO satellite communications. However, the performance comparison with the orthogonal frequency division multiplexing (OFDM) modulation and the resource allocation scheme for multiuser OTFS-based LEO satellite communication system have rarely been investigated. In this paper, we conduct a performance comparison under various channel conditions between the OTFS and OFDM modulations, encompassing evaluations of sum-rate and bit error ratio (BER). Additionally, we investigate the joint optimal allocation of power and delay-Doppler resource blocks aiming at maximizing sum-rate for multiuser downlink OTFS-based LEO satellite communication systems. Unlike the conventional modulations relaying on complex input-output relations within the Time-Frequency (TF) domain, the OTFS modulation exploits both time and frequency diversities, i.e., delay and Doppler shifts remain constant during a OTFS frame, which facilitates a DD domain input-output simple relation for our investigation. We transform the resulting non-convex and combinatorial optimization problem into an equivalent difference of convex problem by decoupling the conditional constraints, and solve the transformed problem via penalty convex-concave procedure algorithm. Simulation results demonstrate that the OTFS modulation is robust to carrier frequency offsets (CFO) caused by high-mobility of LEO satellites, and has superior performance to the OFDM modulation. Moreover, numerical results indicate that our proposed resource allocation scheme has higher sum-rate than existed schemes for the OTFS modulation, such as delay divided multiple access and Doppler divided multiple access, especially in the high signal-to-noise ratio (SNR) regime.

cs.IT↗

arXiv2024

Validity of energy conditions of matter in traversable wormholes under the $f(Q)$ modified gravity theory

In the framework of the theory of general relativity, in order to obtain stable traversable wormholes, matter needs to violate the null energy condition. It is well known that the violation of the energy condition (EC) of matter leads to various physical problems. To address this issue, researchers have turned their attention to exploring modified theories of gravity, aiming to avoid the violation of ECs by introducing geometric terms. In this paper, within the framework of the $f(Q)$ modified gravitational theory, we investigate the effectiveness of ECs for matter in traversable wormholes. We examine the compliance of four types of energy conditions (weak energy condition, null energy condition, dominant energy condition, and strong energy condition) in the model by selecting a power-law model for $f(Q)$ and considering different shape functions $b(r)$. Our study reveals that for traversable wormholes realized through the $f(Q)$ modified gravity theory using the power-law model $f(Q)=a(-Q)^n$, all four types of ECs for matter can be satisfied. There is no need to introduce exotic matter (violating the null energy condition) or special matter (violating other energy conditions) artificially in the physics of wormholes.

gr-qc↗

arXiv2024

CoGenT: A Content-oriented Generative-hit Framework for Content Delivery Networks

The service provided by content delivery networks (CDNs) may overlook content locality, leaving room for potential performance improvement. In this study, we explore the feasibility of leveraging generated data as a replacement for fetching data in missing scenarios based on content locality. Due to sufficient local computing resources and reliable generation efficiency, we propose a content-oriented generative-hit framework (CoGenT) for CDNs. CoGenT utilizes idle computing resources on edge nodes to generate requested data based on similar or related cached data to achieve hits. Our implementation in a real-world system demonstrates that CoGenT reduces the average access latency by half. Additionally, experiments conducted on a simulator also confirm that CoGenT can enhance existing caching algorithms, resulting in reduced latency and bandwidth usage.

cs.DC↗

arXiv2024

High-order topological pumping on a superconducting quantum processor

High-order topological phases of matter refer to the systems of $n$-dimensional bulk with the topology of $m$-th order, exhibiting $(n-m)$-dimensional boundary modes and can be characterized by topological pumping. Here, we experimentally demonstrate two types of second-order topological pumps, forming four 0-dimensional corner localized states on a 4$\times$4 square lattice array of 16 superconducting qubits. The initial ground state of the system for half-filling, as a product of four identical entangled 4-qubit states, is prepared using an adiabatic scheme. During the pumping procedure, we adiabatically modulate the superlattice Bose-Hubbard Hamiltonian by precisely controlling both the hopping strengths and on-site potentials. At the half pumping period, the system evolves to a corner-localized state in a quadrupole configuration. The robustness of the second-order topological pump is also investigated by introducing different on-site disorder. Our work studies the topological properties of high-order topological phases from the dynamical transport picture using superconducting qubits, which would inspire further research on high-order topological phases.

quant-ph↗

arXiv2024

MaskMA: Towards Zero-Shot Multi-Agent Decision Making with Mask-Based Collaborative Learning

Building a single generalist agent with strong zero-shot capability has recently sparked significant advancements. However, extending this capability to multi-agent decision making scenarios presents challenges. Most current works struggle with zero-shot transfer, due to two challenges particular to the multi-agent settings: (a) a mismatch between centralized training and decentralized execution; and (b) difficulties in creating generalizable representations across diverse tasks due to varying agent numbers and action spaces. To overcome these challenges, we propose a Mask-Based collaborative learning framework for Multi-Agent decision making (MaskMA). Firstly, we propose to randomly mask part of the units and collaboratively learn the policies of unmasked units to handle the mismatch. In addition, MaskMA integrates a generalizable action representation by dividing the action space into intrinsic actions solely related to the unit itself and interactive actions involving interactions with other units. This flexibility allows MaskMA to tackle tasks with varying agent numbers and thus different action spaces. Extensive experiments in SMAC reveal MaskMA, with a single model trained on 11 training maps, can achieve an impressive 77.8% average zero-shot win rate on 60 unseen test maps by decentralized execution, while also performing effectively on other types of downstream tasks (e.g., varied policies collaboration, ally malfunction, and ad hoc team play).

cs.AI↗

arXiv2024

DPBalance: Efficient and Fair Privacy Budget Scheduling for Federated Learning as a Service

Federated learning (FL) has emerged as a prevalent distributed machine learning scheme that enables collaborative model training without aggregating raw data. Cloud service providers further embrace Federated Learning as a Service (FLaaS), allowing data analysts to execute their FL training pipelines over differentially-protected data. Due to the intrinsic properties of differential privacy, the enforced privacy level on data blocks can be viewed as a privacy budget that requires careful scheduling to cater to diverse training pipelines. Existing privacy budget scheduling studies prioritize either efficiency or fairness individually. In this paper, we propose DPBalance, a novel privacy budget scheduling mechanism that jointly optimizes both efficiency and fairness. We first develop a comprehensive utility function incorporating data analyst-level dominant shares and FL-specific performance metrics. A sequential allocation mechanism is then designed using the Lagrange multiplier method and effective greedy heuristics. We theoretically prove that DPBalance satisfies Pareto Efficiency, Sharing Incentive, Envy-Freeness, and Weak Strategy Proofness. We also theoretically prove the existence of a fairness-efficiency tradeoff in privacy budgeting. Extensive experiments demonstrate that DPBalance outperforms state-of-the-art solutions, achieving an average efficiency improvement of $1.44\times \sim 3.49 \times$, and an average fairness improvement of $1.37\times \sim 24.32 \times$.

cs.DC↗

arXiv2024

Broadband squeezed light field by magnetostriction in an opto-magnomechanical

We present a novel mechanism for generating a wide bandwidth squeezed optical output field in an opto-magnomechanical system. In this system, the magnon (mechanical) mode in the yttrium-iron-garnet crystal is coupled to the microwave field (optical field) through magnetic dipole (radiation pressure) interaction. The magnetostrictive force induced by the yttrium-iron-garnet crystal causes a mechanical displacement and creates a quadrature squeezed magnon mode. Eventually, this quadrature squeezed mechanical mode is transferred to the output optical field through state-swap interaction. Our results demonstrate the optimal parameter range for obtaining a stable squeezed optical output field with a wide bandwidth. Moreover, the squeezed light field exhibits strong robustness to environmental temperature. The new scheme we propose has potential applications in quantum precision measurements, quantum wireless networks, quantum radar, etc.

quant-ph↗

arXiv2024

Estimating On-road Transportation Carbon Emissions from Open Data of Road Network and Origin-destination Flow Data

Accounting for over 20% of the total carbon emissions, the precise estimation of on-road transportation carbon emissions is crucial for carbon emission monitoring and efficient mitigation policy formulation. However, existing estimation methods typically depend on hard-to-collect individual statistics of vehicle miles traveled to calculate emissions, thereby suffering from high data collection difficulty. To relieve this issue by utilizing the strong pattern recognition of artificial intelligence, we incorporate two sources of open data representative of the transportation demand and capacity factors, the origin-destination (OD) flow data and the road network data, to build a hierarchical heterogeneous graph learning method for on-road carbon emission estimation (HENCE). Specifically, a hierarchical graph consisting of the road network level, community level, and region level is constructed to model the multi-scale road network-based connectivity and travel connection between spatial areas. Heterogeneous graphs consisting of OD links and spatial links are further built at both the community level and region level to capture the intrinsic interactions between travel demand and road network accessibility. Extensive experiments on two large-scale real-world datasets demonstrate HENCE's effectiveness and superiority with R-squared exceeding 0.75 and outperforming baselines by 9.60% on average, validating its success in pioneering the use of artificial intelligence to empower carbon emission management and sustainability development. The implementation codes are available at this link: https://github.com/tsinghua-fib-lab/HENCE.

cs.LG↗

arXiv2024

Exploring the Optimal Cycle for Quantum Heat Engine using Reinforcement Learning

Quantum thermodynamic relationships in emerging nanodevices are significant but often complex to deal with. The application of machine learning in quantum thermodynamics has provided a new perspective. This study employs reinforcement learning to output the optimal cycle of quantum heat engine. Specifically, the soft actor-critic algorithm is adopted to optimize the cycle of three-level coherent quantum heat engine with the aim of maximal average power. The results show that the optimal average output power of the coherent three-level heat engine is 1.28 times greater than the original cycle (steady limit). Meanwhile, the efficiency of the optimal cycle is greater than the Curzon-Ahlborn efficiency as well as reporting by other researchers. Notably, this optimal cycle can be fitted as an Otto-like cycle by applying the Boltzmann function during the compression and expansion processes, which illustrates the effectiveness of the method.

quant-ph↗

arXiv2024

Regulating CPU Temperature With Thermal-Aware Scheduling Using a Reduced Order Learning Thermal Model

Modern real-time systems utilize considerable amounts of power while executing computation-intensive tasks. The execution of these tasks leads to significant power dissipation and heating of the device. It therefore results in severe thermal issues like temperature escalation, high thermal gradients, and excessive hot spot formation, which may result in degrading chip performance, accelerating device aging, and premature failure. Thermal-Aware Scheduling (TAS) enables optimization of thermal dissipation to maintain a safe thermal state. In this work, we implement a new TAS algorithm, POD-TAS, which manages the thermal behavior of a multi-core CPU based on a defined set of states and their transitions. We compare the performances of a dynamic RC thermal circuit simulator (HotSpot) and a reduced order Proper Orthogonal Decomposition (POD)-based thermal model and we select the latter for use in our POD-TAS algorithm. We implement a novel simulation-based evaluation methodology to compare TAS algorithms. This methodology is used to evaluate the performance of the proposed POD-TAS algorithm. Additionally, we compare the performance of a state of the art TAS algorithm, RT-TAS, to our proposed POD-TAS algorithm. Furthermore, we utilize the COMBS benchmark suite to provide CPU workloads for task scheduling. Our experimental results on a multi-core processor using a set of 4 benchmarks demonstrate that the proposed POD-TAS method can improve thermal performance by decreasing the peak thermal variance by 53.0% and the peak chip temperature of 29.01%. Using a set of 8 benchmarks, the comparison of the two algorithms shows a decrease of 29.57% in the peak spatial variance of the chip temperature and 26.26% in the peak chip temperature. We also identify several potential future research directions.

eess.SY↗

arXiv2024

Building Open-Ended Embodied Agent via Language-Policy Bidirectional Adaptation

Building embodied agents on integrating Large Language Models (LLMs) and Reinforcement Learning (RL) have revolutionized human-AI interaction: researchers can now leverage language instructions to plan decision-making for open-ended tasks. However, existing research faces challenges in meeting the requirement of open-endedness. They typically either train LLM/RL models to adapt to a fixed counterpart, limiting exploration of novel skills and hindering the efficacy of human-AI interaction. To this end, we present OpenPAL, a co-training framework comprising two stages: (1) fine-tuning a pre-trained LLM to translate human instructions into goals for planning, and goal-conditioned training a policy for decision-making; (2) co-training to align the LLM and policy, achieving instruction open-endedness. We conducted experiments using Contra, an open-ended FPS game, demonstrating that an agent trained with OpenPAL not only comprehends arbitrary instructions but also exhibits efficient execution. These results suggest that OpenPAL holds the potential to construct open-ended embodied agents in practical scenarios.

cs.AI↗

arXiv2024

Entanglement enhancement of two different magnon modes via nonlinear effect in cavity magnomechanics

We present a scheme to enhance two different magnon modes entanglement in cavity magnomechanics via nonlinear effect. The scheme demonstrated that nonlinear effects enhance entanglement of the two magnon modes. Moreover, the entanglement of the two magnon modes is also significantly enhanced by microwave parametric amplification (PA) and magnon self-Kerr nonlinearity. Not only dose nonlinear effect enhances the strength of entanglement, but it also increases the robustness of entanglement against temperature. Our proposed scheme plays an important role in the research of fundamental theories of quantum physics and quantum information processing theory.

quant-ph↗

arXiv2024

UV-SAM: Adapting Segment Anything Model for Urban Village Identification

Urban villages, defined as informal residential areas in or around urban centers, are characterized by inadequate infrastructures and poor living conditions, closely related to the Sustainable Development Goals (SDGs) on poverty, adequate housing, and sustainable cities. Traditionally, governments heavily depend on field survey methods to monitor the urban villages, which however are time-consuming, labor-intensive, and possibly delayed. Thanks to widely available and timely updated satellite images, recent studies develop computer vision techniques to detect urban villages efficiently. However, existing studies either focus on simple urban village image classification or fail to provide accurate boundary information. To accurately identify urban village boundaries from satellite images, we harness the power of the vision foundation model and adapt the Segment Anything Model (SAM) to urban village segmentation, named UV-SAM. Specifically, UV-SAM first leverages a small-sized semantic segmentation model to produce mixed prompts for urban villages, including mask, bounding box, and image representations, which are then fed into SAM for fine-grained boundary identification. Extensive experimental results on two datasets in China demonstrate that UV-SAM outperforms existing baselines, and identification results over multiple years show that both the number and area of urban villages are decreasing over time, providing deeper insights into the development trends of urban villages and sheds light on the vision foundation models for sustainable cities. The dataset and codes of this study are available at https://github.com/tsinghua-fib-lab/UV-SAM.

cs.CV↗

arXiv2024

StrokeNUWA: Tokenizing Strokes for Vector Graphic Synthesis

To leverage LLMs for visual synthesis, traditional methods convert raster image information into discrete grid tokens through specialized visual modules, while disrupting the model's ability to capture the true semantic representation of visual scenes. This paper posits that an alternative representation of images, vector graphics, can effectively surmount this limitation by enabling a more natural and semantically coherent segmentation of the image information. Thus, we introduce StrokeNUWA, a pioneering work exploring a better visual representation ''stroke tokens'' on vector graphics, which is inherently visual semantics rich, naturally compatible with LLMs, and highly compressed. Equipped with stroke tokens, StrokeNUWA can significantly surpass traditional LLM-based and optimization-based methods across various metrics in the vector graphic generation task. Besides, StrokeNUWA achieves up to a 94x speedup in inference over the speed of prior methods with an exceptional SVG code compression ratio of 6.9%.

cs.CV↗