Searcharxiv⌕ Search

arXiv subjects

Cheng Hua

Publications and source records attributed to Cheng Hua.

At least 19 recordsLinked to original sources

Gaussian Process Bandits with Paired Auxiliary Observations

Bayesian optimization (BO) is widely used for expensive black-box optimization, yet observation noise makes each evaluation less informative and slows posterior learning. Standard BO learns from costly outcomes gathered during optimization. In many applications, however, low-cost auxiliary data are available beforehand. We introduce Auxiliary-Informed Gaussian Process Bandit (AI-GPB), a variance-reduction method that combines an offline estimate of the auxiliary mean with a correction coefficient learned from paired online observations. Notably, it exploits correlation between paired observation errors without requiring the auxiliary mean to approximate the objective. We establish finite-sample GP confidence bounds and regret guarantees, including explicit cumulative and simple regret rates for Matérn kernels. Numerical experiments across continuous BO benchmarks with varied landscapes and dimensions show consistent reductions in cumulative and simple regret relative to vanilla GP-UCB, even when within-pair correlation is moderate and the auxiliary mean is estimated from noisy offline data.

cs.LG↗

Holding Agents Accountable with Verifiable Agent Insurance

While AI agents are useful for many tasks, they regularly misbehave in destructive ways, such as deleting important data. While prior work has focused on preventing such misbehavior, we consider how to recover after an agent misbehaves. We propose verifiable agent insurance (VAI), a framework for agent providers to commit financially to promises, a list of programmatically-specified policies that an agent is meant to follow. These promises are embedded in a smart contract on an immutable ledger (e.g., a blockchain) to prevent tampering, along with a monetary escrow. A user who detects a violation of committed promises can prompt a verifier to review agent execution traces, which are logged through the agent harness's shared tool-execution path, without instrumenting each tool separately. Upon verification of the violation, the agent provider's escrow funds are paid to the reporting user. We instantiate VAI as an end-to-end implementation on an Ethereum layer-2 network, and evaluate its coverage on three prominent agent benchmarks, two of which we re-annotated to obtain task-level policy-compliance labels. Currently, many agents do not emit user authorizations and security-relevant information as explicit, structured data objects -- rather, they embed the data needed to verify policy violations in natural language exchanges. Even so, VAI manages to detect violations in 42.2% of runs containing violations across three benchmarks. By augmenting agents to explicitly emit such objects, violation detection rises to 83.0% of traces. Our findings suggest that for improved auditability, agents should be co-designed with their security policies to emit explicit, structured records of user authorization and security-relevant actions.

cs.CR↗

One Analyst Is Not Ground Truth: Grading Agent-Built Financial Models Against Observed Professional Practice

Language model agents can now construct complete financial models, but it remains unclear whether they have learned the financial judgment that gives those models meaning. Existing benchmarks often anchor correctness to an expert-authored solution (e.g., numerical targets or detailed rubrics) which introduces a implicit assumption: \emph{one expert solution can serve as ground truth.} This is appropriate when finance provides a unique answer, but not when judgment is required. Evidence from professional practice challenges this assumption: When financial models built by different analysts for the same company are graded against one another, the median score is only 0.33 under standard tolerances, \emph{revealing that single-reference grading confounds professional disagreement with error.} We therefore introduce GAUGE, a benchmark that decomposes financial-model evaluation into deterministic checks, rubric judgments and numerical rules. Built from 1,001 professional valuation spanning 922 companies and all 25 GICS industry groups, GAUGE checks mechanical properties deterministically and evaluates judgment-bearing quantities against ranges in professional practice. We then validate GAUGE as a measurement instrument by testing expertise ordering, held-out professional values, and robustness to LM judgements. Our findings show that current LM agents are far better at constructing financial models than at deriving the company-specific assumptions that drive valuation, a gap that even persists after fine-tuning.

cs.LG↗

Alpha-R1: Alpha Screening with LLM Reasoning via Reinforcement Learning

Signal decay and regime shifts pose recurring challenges for data-driven investment strategies in non-stationary markets, where conventional time-series and machine learning approaches often struggle to generalize beyond historical correlations. While large language models (LLMs) offer strong capabilities for processing unstructured information, their potential to support quantitative factor screening through explicit economic reasoning remains underexplored. Existing factor-based methods typically reduce alphas to numerical time series, overlooking the semantic rationale that determines when a factor is economically relevant. We present Alpha-R1, an RL-aligned LLM framework for context-aware alpha screening. Its core mechanism, semantic gating, evaluates each candidate factor's semantic profile against a dynamically constructed market state description, selecting a sparse subset of factors whose economic rationale aligns with current market conditions. The selection model is trained via group relative policy optimization (GRPO), using realized portfolio returns as the primary reward signal. Under a 12-month out-of-sample evaluation, Alpha-R1 achieves annualized returns of 47.87% on S&P 500 and 40.57% on CSI 300 with Sharpe ratios of 1.62 and 2.23. These results, obtained under a bounded candidate-pool evaluation protocol, provide evidence for second-stage semantic factor reranking in non-stationary markets. The full implementation and resources are available at https://github.com/FinStep-AI/Alpha-R1.

q-fin.TR↗

M3TR: Temporal Retrieval Enhanced Multi-Modal Micro-video Popularity Prediction

Accurately predicting the popularity of micro-videos is a critical but challenging task, characterized by volatile, `rollercoaster-like' engagement dynamics. Existing methods often fail to capture these complex temporal patterns, leading to inaccurate long-term forecasts. This failure stems from two fundamental limitations: \ding{172} a superficial understanding of user feedback dynamics, which overlooks the mutually exciting and decaying nature of interactions such as likes, comments, and shares; and~\ding{173} retrieval mechanisms that rely solely on static content similarity, ignoring the crucial patterns of how a video's popularity evolves over time. To address these limitations, we propose \textbf{M$^3$TR}, a \textbf{T}emporal \textbf{R}etrieval enhanced \textbf{M}ulti-\textbf{M}odal framework that uniquely synergizes fine-grained temporal modeling with a novel temporal-aware retrieval process for \textbf{M}icro-video popularity prediction. At its core, M$^3$TR introduces a Mamba-Hawkes Process (MHP) module to explicitly model user feedback as a sequence of self-exciting events, capturing the intricate, long-range dependencies within user interactions (for \textbf{limitation} \ding{172}). This rich temporal representation then powers a temporal-aware retrieval engine that identifies historically relevant videos based on a combined similarity of both their multi-modal content (visual, audio, text) and their popularity trajectories (for \textbf{limitation} \ding{173}). By augmenting the target video's features with this retrieved knowledge, M$^3$TR achieves a comprehensive understanding of prediction. Extensive experiments on two real-world datasets demonstrate the superiority of our framework. M$^3$TR achieves state-of-the-art performance, outperforming previous methods by up to \textbf{19.3}\% in nMSE and showing significant gains in addressing long-term prediction challenges.

cs.MM↗

DASH: Decoupled Adaptive Surrogate - Acquisition Harness for Automated Bayesian Optimization

Bayesian optimization (BO) relies on a surrogate model and an acquisition function, yet the most suitable choices vary across tasks and optimization stages. Automated Bayesian optimization (AutoBO) addresses this variability by adapting BO components online. However, existing AutoBO methods either adapt one component, leaving the other mismatched and creating a bottleneck, or jointly select surrogate--acquisition pairs under a shared criterion, overlooking their distinct roles: surrogate selection depends on predictive reliability, whereas acquisition adaptation should respond to campaign context.In this paper, we propose DASH, a Decoupled Adaptive Surrogate--Acquisition Harness for large-language- model (LLM)-enhanced AutoBO. DASH selects surrogates by predictive reliability, uncertainty calibration, and ranking consistency; its two-stage acquisition controller periodically reallocates quotas across acquisition functions, builds a BO shortlist accordingly, and delegates final selection to an LLM. DASH also incorporates an integrated harness, consisting of knowledge-guided warm start and structured memory, to ground optimization in domain knowledge and accumulated feedback. Across four chemical optimization tasks, DASH outperforms the best AutoBO baseline by 12.51% in trajectory-level Acceleration Factor and 5.00% in endpoint Enhancement Factor. Results remain strong across LLM backbones, and ablations verify the complementary contributions of all components. Full-table and behavioral contamination checks find no detectable evidence that direct benchmark memorization or source-cell leakage explains these gains.

cs.AI↗

Neo-Classic: A Benchmark for Evaluating Linguistic-Aesthetic Reasoning in Classical Chinese Poetry

While Large Language Models (LLMs) achieve high accuracy on established Classical Chinese Poetry benchmarks, it remains challenging to distinguish transferable Linguistic-Aesthetic Reasoning from reliance on familiar pre-training patterns. To address this issue, we introduce Neo-Classic, an evaluation benchmark that combines a constructionist Out-of-Sample (OOS) dataset with a suite of reverse understanding probes. Unlike traditional benchmarks that rely on verification or generation over historical corpora, Neo-Classic comprises strictly metrical poetry authored by contemporary experts, reducing the possibility of direct retrieval. We evaluate state-of-the-art models, including Qwen3-Max, Gemini-3-Pro, and DeepSeek-V3.2, across five behavioral probes designed to test hierarchical constraint satisfaction. Our results reveal two primary limitations. First, a performance gap of 20 to 50 percent emerges when models transition from historical to contemporary texts. Second, models exhibit substantial difficulties in discourse-level ordering tasks, with standard accuracy remaining low (0 to 13 percent). Although expert-level guidance improves the performance of reasoning-enhanced models to 36 percent, a notable gap with human experts persists. These findings suggest that while current LLMs capture local formal patterns, they struggle with global hierarchical planning required for robust Linguistic-Aesthetic Reasoning.

cs.CL↗

ALAS: Additive Learnable Alpha-Stable Kernels for Flexible Bayesian Optimization

Bayesian Optimization is widely used for expensive black-box optimization, yet its success often depends on choosing a kernel that matches the objective's unknown structure. In this work, we propose ALAS, a flexible Gaussian Process kernel family built from symmetric $α$-stable spectral components. By learning the stability parameter $α$, ALAS adapts its effective smoothness from data, capturing both smooth trends and sharp irregularities. We present two parameterizations: ALAS, a single stationary component with joint spectral modulation, and ALAS-Sep, a separable variant that learns dimension-wise tail behavior to improve robustness on approximately decomposable objectives. Experiments on standard benchmarks and real-world surrogates demonstrate strong and robust performance across diverse settings.

cs.LG↗

On-Demand Service Zone Design for Energy-Constrained Spatial Queueing Systems

Electric service vehicles (ESVs), such as mobile chargers and drone-based service units, are becoming an important operational resource for on-demand service systems. Unlike conventional spatial servers, ESV operations are shaped by battery limits and recharging needs, which affect dispatch feasibility and spatial deployment decisions. We develop an energy-constrained hypercube spatial queueing model that embeds battery-state dynamics into the classical hypercube framework and uses a semi-Markov representation to estimate steady-state performance. We then formulate a joint location--zoning problem for station placement and service zone design. The resulting large-scale mixed-integer nonlinear program admits a set partitioning reformulation whose column coefficients are not available in closed form. We therefore develop a Branch-Price-and-Evaluation framework for set partitioning problems with externally computable column coefficients: upper-bounding surrogates guide pricing, and iterative exact evaluation updates the coefficients of active columns. Computational results show that explicit energy modeling significantly reduces false service promises and yields more credible planning decisions. They also reveal a load-dependent reversal in zoning: pooling is preferable under light demand, whereas tighter zoning becomes more profitable as demand increases. Over the tested range, profitability is driven more by zoning than by battery improvement, suggesting that managers should get service zone design right before investing in battery upgrades; this caution is reinforced by the counterintuitive finding that larger batteries may delay replenishment and reduce fleet readiness under sparse demand. These findings show that energy feasibility is not merely a matter of battery-capacity expansion, but a design dimension that shapes service-zone configuration.

cs.PF↗

CatBOX: A Categorical-Continuous Bayesian Optimization with Spectral Mixture Kernels for Accelerated Catalysis Experiments

Identifying optimal catalyst compositions and reaction conditions is central in catalysis research, yet remains challenging due to the vast multidimensional design spaces encompassing both continuous and categorical parameters. In this work, we present CatBOX, a Bayesian Optimization method for accelerated catalytic experimental design that jointly optimizes categorical and continuous experimental parameters. Our approach introduces a novel spectral mixture kernel that combines the inverse Fourier transform of Gaussian and Cauchy mixtures to provide a flexible representation of the continuous parameter space, capturing both smooth and non-smooth variations. Categorical choices, such as catalyst compositions and support types, are navigated via trust regions based on Hamming distance. For performance evaluation, CatBOX was theoretically verified based on information theory and benchmarked on a series of synthetic functions, achieving more than a 3-fold improvement relative to the best-performing baseline and a 19-fold improvement relative to random search on average across tasks. Additionally, three real-life catalytic experiments, including oxidative coupling of methane, urea-selective catalytic reduction, and direct arylation of imidazoles, were further used for in silico benchmarking, where CatBOX reliably identified top catalyst recipes and reaction conditions with the highest efficiencies in the absence of any a priori knowledge. Finally, we develop an open-source, code-free online platform to facilitate trial deployment in real experimental settings, particularly for self-driving laboratory environments.

cs.LG↗

A Geometrically Convergent Solution to Spatial Hypercube Queueing Models

The hypercube queueing model was initially developed to address spatial queueing problems and has found wide applications in emergency services, such as ambulance and police systems. While the model was originally designed for homogeneous service rates, we extend it to handle heterogeneous service rates by devising an exact solution through a birth-death process and an equivalent reformulation. We demonstrate that our algorithm converges to the exact solution at a geometric rate. Additionally, we developed a parallel algorithm that leverages the convergence property and two structural features of the hypercube model, achieving more than 91% parallelization. Numerical experiments on emergency medical service systems show that our sequential algorithm is over 1,000 times faster than the sparse solver and more than 500 times faster than discrete-event simulation, while maintaining high accuracy. The parallel algorithm further improves efficiency, achieving an approximately eightfold speedup with 12 processing units, with additional gains possible when more computational resources are available. Overall, the proposed algorithms improve computational efficiency and enable the solution of large-scale problems that are otherwise intractable using traditional approaches.

math.OC↗

Trade-R1: Bridging Verifiable Rewards to Stochastic Environments via Process-Level Reasoning Verification

Reinforcement Learning (RL) has enabled Large Language Models (LLMs) to achieve remarkable reasoning in domains like mathematics and coding, where verifiable rewards provide clear signals. However, extending this paradigm to financial decision is challenged by the market's stochastic nature: rewards are verifiable but inherently noisy, causing standard RL to degenerate into reward hacking. To address this, we propose Trade-R1, a model training framework that bridges verifiable rewards to stochastic environments via process-level reasoning verification. Our key innovation is a verification method that transforms the problem of evaluating reasoning over lengthy financial documents into a structured Retrieval-Augmented Generation (RAG) task. We construct a triangular consistency metric, assessing pairwise alignment between retrieved evidence, reasoning chains, and decisions to serve as a validity filter for noisy market returns. We explore two reward integration strategies: Fixed-effect Semantic Reward (FSR) for stable alignment signals, and Dynamic-effect Semantic Reward (DSR) for coupled magnitude optimization. Experiments on different country asset selection demonstrate that our paradigm reduces reward hacking, with DSR achieving superior cross-market generalization while maintaining the highest reasoning consistency.

cs.AI↗

Optimizing Server Locations in Spatial Queues: Parametric and Nonparametric Bayesian Optimization

This paper presents a new model for solving the optimal server location problem in a spatial hypercube queueing model. Unlike deterministic location models, our approach accounts for server availability, varying utilization levels, and dependencies across servers. We prove that the problem is NP-hard and establish lower and upper bounds, as well as asymptotic results, by relating it to special cases of the classical $p$-Median problem. To address the computational challenge, we propose two Bayesian optimization approaches: (i) a parametric approach based on a sparse Bayesian linear model with second-order interactions, and (ii) a nonparametric approach using a Gaussian process surrogate with the $p$-Median objective as the prior mean function. We prove that both methods achieve sublinear regret and converge to the optimal solution. Numerical experiments and a case study using real-world data from the St. Paul, Minnesota, emergency response system show that our approaches consistently identify optimal solutions and outperform all baseline methods.

math.OC↗

SPOT: Scalable Policy Optimization with Trees for Markov Decision Processes

Interpretable reinforcement learning policies are essential for high-stakes decision-making, yet optimizing decision tree policies in Markov Decision Processes (MDPs) remains challenging. We propose SPOT, a novel method for computing decision tree policies, which formulates the optimization problem as a mixed-integer linear program (MILP). To enhance efficiency, we employ a reduced-space branch-and-bound approach that decouples the MDP dynamics from tree-structure constraints, enabling efficient parallel search. This significantly improves runtime and scalability compared to previous methods. Our approach ensures that each iteration yields the optimal decision tree. Experimental results on standard benchmarks demonstrate that SPOT achieves substantial speedup and scales to larger MDPs with a significantly higher number of states. The resulting decision tree policies are interpretable and compact, maintaining transparency without compromising performance. These results demonstrate that our approach simultaneously achieves interpretability and scalability, delivering high-quality policies an order of magnitude faster than existing approaches.

cs.LG↗

Identifying All ε-Best Arms in (Misspecified) Linear Bandits

Motivated by the need to efficiently identify multiple candidates in high trial-and-error cost tasks such as drug discovery, we propose a near-optimal algorithm to identify all ε-best arms (i.e., those at most ε worse than the optimum). Specifically, we introduce LinFACT, an algorithm designed to optimize the identification of all ε-best arms in linear bandits. We establish a novel information-theoretic lower bound on the sample complexity of this problem and demonstrate that LinFACT achieves instance optimality by matching this lower bound up to a logarithmic factor. A key ingredient of our proof is to integrate the lower bound directly into the scaling process for upper bound derivation, determining the termination round and thus the sample complexity. We also extend our analysis to settings with model misspecification and generalized linear models. Numerical experiments, including synthetic and real drug discovery data, demonstrate that LinFACT identifies more promising candidates with reduced sample complexity, offering significant computational efficiency and accelerating early-stage exploratory experiments.

stat.ML↗

Multi-modal and Metadata Capture Model for Micro Video Popularity Prediction

As short videos have become the primary form of content consumption across various industries, accurately predicting their popularity has become key to enhancing user engagement and optimizing business strategies. This report presents a solution for the 2024 INFORMS Data Mining Challenge, focusing on our developed 3M model (Multi-modal and Metadata Capture Model), which is a multi-modal popularity prediction model. The 3M model integrates video, audio, descriptions, and metadata to fully explore the multidimensional information of short videos. We employ a retriever-based method to retrieve relevant instances from a multi-modal memory bank, filtering similar videos based on visual, acoustic, and text-based features for prediction. Additionally, we apply a random masking method combined with a semi-supervised model for incomplete multi-modalities to leverage the metadata of videos. Ultimately, we use a network to synthesize both approaches, significantly improving the accuracy of predictions. Compared to traditional tag-based algorithms, our model outperforms existing methods on the validation set, showing a notable increase in prediction accuracy. Our research not only offers a new perspective on understanding the drivers of short video popularity but also provides valuable data support for identifying market opportunities, optimizing advertising strategies, and enhancing content creation. We believe that the innovative methodology proposed in this report provides practical tools and valuable insights for professionals in the field of short video popularity prediction, helping them effectively address future challenges.

cs.MM↗

Reinforcement-Learning Portfolio Allocation with Dynamic Embedding of Market Information

We develop a portfolio allocation framework that leverages deep learning techniques to address challenges arising from high-dimensional, non-stationary, and low-signal-to-noise market information. Our approach includes a dynamic embedding method that reduces the non-stationary, high-dimensional state space into a lower-dimensional representation. We design a reinforcement learning (RL) framework that integrates generative autoencoders and online meta-learning to dynamically embed market information, enabling the RL agent to focus on the most impactful parts of the state space for portfolio allocation decisions. Empirical analysis based on the top 500 U.S. stocks demonstrates that our framework outperforms common portfolio benchmarks and the predict-then-optimize (PTO) approach using machine learning, particularly during periods of market stress. Traditional factor models do not fully explain this superior performance. The framework's ability to time volatility reduces its market exposure during turbulent times. Ablation studies confirm the robustness of this performance across various reinforcement learning algorithms. Additionally, the embedding and meta-learning techniques effectively manage the complexities of high-dimensional, noisy, and non-stationary financial data, enhancing both portfolio performance and risk management.

q-fin.PM↗

Collaborative Intelligence in Sequential Experiments: A Human-in-the-Loop Framework for Drug Discovery

Drug discovery is a complex process that involves sequentially screening and examining a vast array of molecules to identify those with the target properties. This process, also referred to as sequential experimentation, faces challenges due to the vast search space, the rarity of target molecules, and constraints imposed by limited data and experimental budgets. To address these challenges, we introduce a human-in-the-loop framework for sequential experiments in drug discovery. This collaborative approach combines human expert knowledge with deep learning algorithms, enhancing the discovery of target molecules within a specified experimental budget. The proposed algorithm processes experimental data to recommend both promising molecules and those that could improve its performance to human experts. Human experts retain the final decision-making authority based on these recommendations and their domain expertise, including the ability to override algorithmic recommendations. We applied our method to drug discovery tasks using real-world data and found that it consistently outperforms all baseline methods, including those which rely solely on human or algorithmic input. This demonstrates the complementarity between human experts and the algorithm. Our results provide key insights into the levels of humans' domain knowledge, the importance of meta-knowledge, and effective work delegation strategies. Our findings suggest that such a framework can significantly accelerate the development of new vaccines and drugs by leveraging the best of both human and artificial intelligence.

cs.AI↗