arXiv subjects

Peng Yang

Publications and source records attributed to Peng Yang.

At least 91 records · Page 5Linked to original sources

arXiv2025

Rationale-guided Prompting for Knowledge-based Visual Question Answering

Recently, Large Language Models (LLMs) have been used for knowledge-based Visual Question Answering (VQA). Despite the encouraging results of previous studies, prior methods prompt LLMs to predict answers directly, neglecting intermediate thought processes. We argue that prior methods do not sufficiently activate the capacities of LLMs. We propose a framework called PLRH that Prompts LLMs with Rationale Heuristics for knowledge-based VQA. The PLRH prompts LLMs with Chain of Thought (CoT) to generate rationale heuristics, i.e., intermediate thought processes, and then leverages the rationale heuristics to inspire LLMs to predict answers. Experiments show that our approach outperforms the existing baselines by more than 2.2 and 2.1 on OK-VQA and A-OKVQA, respectively.

cs.CL↗

arXiv2025

Multi-Agents Based on Large Language Models for Knowledge-based Visual Question Answering

Large Language Models (LLMs) have achieved impressive results in knowledge-based Visual Question Answering (VQA). However existing methods still have challenges: the inability to use external tools autonomously, and the inability to work in teams. Humans tend to know whether they need to use external tools when they encounter a new question, e.g., they tend to be able to give a direct answer to a familiar question, whereas they tend to use tools such as search engines when they encounter an unfamiliar question. In addition, humans also tend to collaborate and discuss with others to get better answers. Inspired by this, we propose the multi-agent voting framework. We design three LLM-based agents that simulate different levels of staff in a team, and assign the available tools according to the levels. Each agent provides the corresponding answer, and finally all the answers provided by the agents are voted to get the final answer. Experiments on OK-VQA and A-OKVQA show that our approach outperforms other baselines by 2.2 and 1.0, respectively.

cs.CL↗

arXiv2025

Learning from Expert Factors: Trajectory-level Reward Shaping for Formulaic Alpha Mining

Reinforcement learning (RL) has successfully automated the complex process of mining formulaic alpha factors, for creating interpretable and profitable investment strategies. However, existing methods are hampered by the sparse rewards given the underlying Markov Decision Process. This inefficiency limits the exploration of the vast symbolic search space and destabilizes the training process. To address this, Trajectory-level Reward Shaping (TLRS), a novel reward shaping method, is proposed. TLRS provides dense, intermediate rewards by measuring the subsequence-level similarity between partially generated expressions and a set of expert-designed formulas. Furthermore, a reward centering mechanism is introduced to reduce training variance. Extensive experiments on six major Chinese and U.S. stock indices show that TLRS significantly improves the predictive power of mined factors, boosting the Rank Information Coefficient by 9.29% over existing potential-based shaping algorithms. Notably, TLRS achieves a major leap in computational efficiency by reducing its time complexity with respect to the feature dimension from linear to constant, which is a significant improvement over distance-based baselines.

cs.LG↗

arXiv2025

Federated Learning in Active STARS-Aided Uplink Networks

Active simultaneously transmitting and reflecting surfaces (ASTARS) have attracted growing research interest due to its ability to alleviate multiplicative fading and reshape the electromagnetic environment across the entire space. In this paper, we utilise ASTARS to assist the federated learning (FL) uplink model transfer and further reduce the number of uploaded parameter counts through over-the-air (OTA) computing techniques. The impact of model aggregation errors on ASTARS-aided FL uplink networks is characterized. We derive an upper bound on the aggregation error of the OTA-FL model and quantify the training loss due to communication errors. Then, we define the performance of OTA-FL as a joint optimization problem that encompasses both the assignment of received beams and the phase shifting of ASTARS, aiming to achieve the maximum learning efficiency and high-quality signal transmission. Numerical results demonstrate that: i) The FL accuracy in ASTARS uplink networks are enhanced compared to that in state-of-the-art networks; ii) The ASTARS enabled FL system achieves the better learning accuracy using fewer active units than other baseline, especially when the dataset is more discrete; and iii) FL accuracy improves with higher amplification power, but excessive amplification makes thermal noise the dominant source of error.

eess.SP↗

arXiv2025

Alleviating Non-identifiability: a High-fidelity Calibration Objective for Financial Market Simulation with Multivariate Time Series Data

The non-identifiability issue has been frequently reported in social simulation works, where different parameters of an agent-based simulation model yield indistinguishable simulated time series data under certain discrepancy metrics. This issue largely undermines the simulation fidelity yet lacks dedicated investigations. This paper theoretically demonstrates that incorporating multiple time series data features during the model calibration phase can exponentially alleviate non-identifiability as the number of features increases. To implement this theoretical finding, a maximization-based aggregation function is proposed based on existing discrepancy metrics to form a new calibration objective function. For verification, the task of calibrating the Financial Market Simulation (FMS), a typical yet complex social simulation, is considered. Empirical studies confirm the significant improvements in alleviating the non-identifiability of calibration tasks. Furthermore, as a model-agnostic method, it achieves much higher simulation fidelity of the chosen FMS model on both synthetic and real market data. Moreover, it is both theoretically and empirically analyzed that as long as the features are selected and not linearly correlated, they can contribute to alleviation, which demonstrates the robustness of the proposed objective. Hence, this work is expected to provide not only a rigorous understanding of non-identifiability in social simulation but also an off-the-shelf high-fidelity calibration objective function for FMS.

q-fin.CP↗

arXiv2025

QuantFactor REINFORCE: Mining Steady Formulaic Alpha Factors with Variance-bounded REINFORCE

Alpha factor mining aims to discover investment signals from the historical financial market data, which can be used to predict asset returns and gain excess profits. Powerful deep learning methods for alpha factor mining lack interpretability, making them unacceptable in the risk-sensitive real markets. Formulaic alpha factors are preferred for their interpretability, while the search space is complex and powerful explorative methods are urged. Recently, a promising framework is proposed for generating formulaic alpha factors using deep reinforcement learning, and quickly gained research focuses from both academia and industries. This paper first argues that the originally employed policy training method, i.e., Proximal Policy Optimization (PPO), faces several important issues in the context of alpha factors mining. Herein, a novel reinforcement learning algorithm based on the well-known REINFORCE algorithm is proposed. REINFORCE employs Monte Carlo sampling to estimate the policy gradient-yielding unbiased but high variance estimates. The minimal environmental variability inherent in the underlying state transition function, which adheres to the Dirac distribution, can help alleviate this high variance issue, making REINFORCE algorithm more appropriate than PPO. A new dedicated baseline is designed to theoretically reduce the commonly suffered high variance of REINFORCE. Moreover, the information ratio is introduced as a reward shaping mechanism to encourage the generation of steady alpha factors that can better adapt to changes in market volatility. Evaluations on real assets data indicate the proposed algorithm boosts correlation with returns by 3.83\%, and a stronger ability to obtain excess returns compared to the latest alpha factors mining methods, which meets the theoretical results well.

q-fin.CP↗

arXiv2025

SimLOB: Learning Representations of Limited Order Book for Financial Market Simulation

Financial market simulation (FMS) serves as a promising tool for understanding market anomalies and the underlying trading behaviors. To ensure high-fidelity simulations, it is crucial to calibrate the FMS model for generating data closely resembling the observed market data. Previous efforts primarily focused on calibrating the mid-price data, leading to essential information loss of the market activities and thus biasing the calibrated model. The Limit Order Book (LOB) data is the fundamental data fully capturing the market micro-structure and is adopted by worldwide exchanges. However, LOB is not applicable to existing calibration objective functions due to its tabular structure not suitable for the vectorized input requirement. This paper proposes to explicitly learn the vectorized representations of LOB with a Transformer-based autoencoder. Then the latent vector, which captures the major information of LOB, can be applied for calibration. Extensive experiments show that the learned latent representation not only preserves the non-linear auto-correlation in the temporal axis, but the precedence between successive price levels of LOB. Besides, it is verified that the performance of the representation learning stage is consistent with the downstream calibration tasks. Thus, this work also progresses the FMS on LOB data, for the first time.

cs.CE↗

arXiv2025

Diversity from Human Feedback

Diversity plays a significant role in many problems, such as ensemble learning, reinforcement learning, and combinatorial optimization. How to define the diversity measure is a longstanding problem. Many methods rely on expert experience to define a proper behavior space and then obtain the diversity measure, which is, however, challenging in many scenarios. In this paper, we propose the problem of learning a behavior space from human feedback and present a general method called Diversity from Human Feedback (DivHF) to solve it. DivHF learns a behavior descriptor consistent with human preference by querying human feedback. The learned behavior descriptor can be combined with any distance measure to define a diversity measure. We demonstrate the effectiveness of DivHF by integrating it with the Quality-Diversity optimization algorithm MAP-Elites and conducting experiments on the QDax suite. The results show that the behavior learned by DivHF is much more consistent with human requirements than the one learned by direct data-driven approaches without human feedback, and makes the final solutions more diverse under human preference. Our contributions include formulating the problem, proposing the DivHF method, and demonstrating its effectiveness through experiments.

cs.LG↗

arXiv2025

Surrogate-Assisted Evolutionary Reinforcement Learning Based on Autoencoder and Hyperbolic Neural Network

Evolutionary Reinforcement Learning (ERL), training the Reinforcement Learning (RL) policies with Evolutionary Algorithms (EAs), have demonstrated enhanced exploration capabilities and greater robustness than using traditional policy gradient. However, ERL suffers from the high computational costs and low search efficiency, as EAs require evaluating numerous candidate policies with expensive simulations, many of which are ineffective and do not contribute meaningfully to the training. One intuitive way to reduce the ineffective evaluations is to adopt the surrogates. Unfortunately, existing ERL policies are often modeled as deep neural networks (DNNs) and thus naturally represented as high-dimensional vectors containing millions of weights, which makes the building of effective surrogates for ERL policies extremely challenging. This paper proposes a novel surrogate-assisted ERL that integrates Autoencoders (AE) and Hyperbolic Neural Networks (HNN). Specifically, AE compresses high-dimensional policies into low-dimensional representations while extracting key features as the inputs for the surrogate. HNN, functioning as a classification-based surrogate model, can learn complex nonlinear relationships from sampled data and enable more accurate pre-selection of the sampled policies without real evaluations. The experiments on 10 Atari and 4 Mujoco games have verified that the proposed method outperforms previous approaches significantly. The search trajectories guided by AE and HNN are also visually demonstrated to be more effective, in terms of both exploration and convergence. This paper not only presents the first learnable policy embedding and surrogate-modeling modules for high-dimensional ERL policies, but also empirically reveals when and why they can be successful.

cs.LG↗

arXiv2025

Survey on Near-Space Information Networks: Channel Modeling, Transmission, and Networking Perspectives

Near-space information networks (NSINs) composed of high-altitude platforms (HAPs) and high- and low-altitude unmanned aerial vehicles (UAVs) are a new regime for providing quick, robust, and cost-efficient sensing and communication services. Precipitated by innovations and breakthroughs in manufacturing, materials, communications, electronics, and control techniques, NSINs have been envisioned as an essential component of the emerging sixth-generation of mobile communication systems. This article reveals some critical issues needing to be tackled in NSINs through conducting experiments and discusses the latest advances in NSINs in the research areas of channel modeling, networking, and transmission from a forward-looking, comparative, and technical evolutionary perspective. In this article, we highlight the characteristics of NSINs and present the promising use cases of NSINs. The impact of airborne platforms' unstable movements on the phase delays of onboard antenna arrays with diverse structures is mathematically analyzed. The recent advances in HAP channel modeling are elaborated on, along with the significant differences between HAP and UAV channel modeling. A comprehensive review of the networking techniques of NSINs in network deployment, handoff management, and network management aspects is provided. Besides, the promising techniques and communication protocols of the physical (PHY) layer, medium access control (MAC) layer, network layer, and transport layer of NSINs for achieving efficient transmission over NSINs are reviewed, and we have conducted experiments with practical NSINs to verify the performance of some techniques. Finally, we outline some open issues and promising directions for NSINs deserved for future study and discuss the corresponding challenges.

cs.NI↗

arXiv2025

Representation Learning of Limit Order Book: A Comprehensive Study and Benchmarking

The Limit Order Book (LOB), the mostly fundamental data of the financial market, provides a fine-grained view of market dynamics while poses significant challenges in dealing with the esteemed deep models due to its strong autocorrelation, cross-feature constrains, and feature scale disparity. Existing approaches often tightly couple representation learning with specific downstream tasks in an end-to-end manner, failed to analyze the learned representations individually and explicitly, limiting their reusability and generalization. This paper conducts the first systematic comparative study of LOB representation learning, aiming to identify the effective way of extracting transferable, compact features that capture essential LOB properties. We introduce LOBench, a standardized benchmark with real China A-share market data, offering curated datasets, unified preprocessing, consistent evaluation metrics, and strong baselines. Extensive experiments validate the sufficiency and necessity of LOB representations for various downstream tasks and highlight their advantages over both the traditional task-specific end-to-end models and the advanced representation learning models for general time series. Our work establishes a reproducible framework and provides clear guidelines for future research. Datasets and code will be publicly available at https://github.com/financial-simulation-lab/LOBench.

cs.CE↗

arXiv2025

Towards Calibrating Financial Market Simulators with High-frequency Data

The fidelity of financial market simulation is restricted by the so-called "non-identifiability" difficulty when calibrating high-frequency data. This paper first analyzes the inherent loss of data information in this difficulty, and proposes to use the Kolmogorov-Smirnov test (K-S) as the objective function for high-frequency calibration. Empirical studies verify that K-S has better identifiability of calibrating high-frequency data, while also leads to a much harder multi-modal landscape in the calibration space. To this end, we propose the adaptive stochastic ranking based negatively correlated search algorithm for improving the balance between exploration and exploitation. Experimental results on both simulated data and real market data demonstrate that the proposed method can obtain up to 36.0% improvement in high-frequency data calibration problems over the compared methods.

cs.CE↗

arXiv2025

Universal critical behavior of generaliezd susceptibilities of net-baryon number at small quark mass

In the limit of small quark masses, the angle between the temperature axis and the applied magnetic field direction in the three-dimensional Ising model vanishes as $m_q^{2/5}$ when mapped onto the QCD $T-μ_B$ phase plane. By selecting two distinct small angles and projecting the Ising model results onto QCD, we have investigated the universal critical behavior of the sixth-, eighth-, and tenth-order susceptibilities of the net-baryon number. When considering only the leading critical contribution, the negative dip in the $μ_B$ dependence of the generalized susceptibilities is not universal, in contrast to the observation in the case where the angle is $90^{\circ}$. Its existence depends on the mapping parameters and the distance to the phase transition line. After incorporating the sub-leading critical contribution, the negative dip is enhanced to some extent but remains a non-robust feature. In contrast, the positive peak structure persists in all cases and represents a robust characteristic of generalized susceptibilities of the net-baryon number near the critical point.

nucl-th↗

arXiv2025

Compression Metadata-assisted RoI Extraction and Adaptive Inference for Efficient Video Analytics

Video analytics demand substantial computing resources, posing significant challenges in computing resource-constrained environment. In this paper, to achieve high accuracy with acceptable computational workload, we propose a cost-effective regions of interest (RoIs) extraction and adaptive inference scheme based on the informative encoding metadata. Specifically, to achieve efficient RoI-based analytics, we explore motion vectors from encoding metadata to identify RoIs in non-reference frames through morphological opening operation. Furthermore, considering the content variation of RoIs, which calls for inference by models with distinct size, we measure RoI complexity based on the bitrate allocation information from encoding metadata. Finally, we design an algorithm that prioritizes scheduling RoIs to models of the appropriate complexity, balancing accuracy and latency. Extensive experimental results show that our proposed scheme reduces latency by nearly 40% and improves 2.2% on average in accuracy, outperforming the latest benchmarks.

cs.MM↗

arXiv2025

Stochastic Calculus and Hochschild Homology

This paper is a case study of probabilistic approach to homological aspects of topological quantum field theory via the example of topological quantum mechanics. We propose topological correlations in terms of large variance limit. An investigation on the relation between probabilistic topological correlations on the circle and Hochschild homology is illustrated.

math.PR↗

arXiv2025

A Survey on Point-of-Interest Recommendation: Models, Architectures, and Security

The widespread adoption of smartphones and Location-Based Social Networks has led to a massive influx of spatio-temporal data, creating unparalleled opportunities for enhancing Point-of-Interest (POI) recommendation systems. These advanced POI systems are crucial for enriching user experiences, enabling personalized interactions, and optimizing decision-making processes in the digital landscape. However, existing surveys tend to focus on traditional approaches and few of them delve into cutting-edge developments, emerging architectures, as well as security considerations in POI recommendations. To address this gap, our survey stands out by offering a comprehensive, up-to-date review of POI recommendation systems, covering advancements in models, architectures, and security aspects. We systematically examine the transition from traditional models to advanced techniques such as large language models. Additionally, we explore the architectural evolution from centralized to decentralized and federated learning systems, highlighting the improvements in scalability and privacy. Furthermore, we address the increasing importance of security, examining potential vulnerabilities and privacy-preserving approaches. Our taxonomy provides a structured overview of the current state of POI recommendation, while we also identify promising directions for future research in this rapidly advancing field.

cs.IR↗

arXiv2025

Brain-Inspired Decentralized Satellite Learning in Space Computing Power Networks

Satellite networks are able to collect massive space information with advanced remote sensing technologies, which is essential for real-time applications such as natural disaster monitoring. However, traditional centralized processing by the ground server incurs a severe timeliness issue caused by the transmission bottleneck of raw data. To this end, Space Computing Power Networks (Space-CPN) emerges as a promising architecture to coordinate the computing capability of satellites and enable on board data processing. Nevertheless, due to the natural limitations of solar panels, satellite power system is difficult to meet the energy requirements for ever-increasing intelligent computation tasks of artificial neural networks. To tackle this issue, we propose to employ spiking neural networks (SNNs), which is supported by the neuromorphic computing architecture, for on-board data processing. The extreme sparsity in its computation enables a high energy efficiency. Furthermore, to achieve effective training of these on-board models, we put forward a decentralized neuromorphic learning framework, where a communication-efficient inter-plane model aggregation method is developed with the inspiration from RelaySum. We provide a theoretical analysis to characterize the convergence behavior of the proposed algorithm, which reveals a network diameter related convergence speed. We then formulate a minimum diameter spanning tree problem on the inter-plane connectivity topology and solve it to further improve the learning performance. Extensive experiments are conducted to evaluate the superiority of the proposed method over benchmarks.

cs.LG↗

arXiv2025

Context-aware Diversity Enhancement for Neural Multi-Objective Combinatorial Optimization

Multi-objective combinatorial optimization (MOCO) problems are prevalent in various real-world applications. Most existing neural MOCO methods rely on problem decomposition to transform an MOCO problem into a series of singe-objective combinatorial optimization (SOCO) problems and train attention models based on a single-step and deterministic greedy rollout. However, inappropriate decomposition and undesirable short-sighted behaviors of previous methods tend to induce a decline in diversity. To address the above limitation, we design a Context-aware Diversity Enhancement algorithm named CDE, which casts the neural MOCO problems as conditional sequence modeling via autoregression (node-level context awareness) and establishes a direct relationship between the mapping of preferences and diversity indicator of reward based on hypervolume expectation maximization (solution-level context awareness). Based on the solution-level context awareness, we further propose a hypervolume residual update strategy to enable the Pareto attention model to capture both local and non-local information of the Pareto set/front. The proposed CDE can effectively and efficiently grasp the context information, resulting in diversity enhancement. Experimental results on three classic MOCO problems demonstrate that our CDE outperforms several state-of-the-art baselines.

cs.LG↗