arXiv subjects

Yuanzhang Xiao

Publications and source records attributed to Yuanzhang Xiao.

At least 19 recordsLinked to original sources

arXiv2026

Fast Generalization after Interpolation via Critically Damped Momentum Optimization

A central problem in machine learning is that models can achieve near-perfect training performance while generalizing substantially less well to unseen examples. This gap is especially acute in high-dimensional, low-sample regimes, where many interpolating solutions exist and optimization must implicitly select among minima with different generalization properties. Following recent theoretical advances on optimization dynamics near the interpolation threshold, we note that the two-regime structure of risk minimization, with loss minimization followed by complexity minimization, motivates a biphasic optimization schedule. We thus theoretically demonstrate that GROKtimizer, a biphasic strategy that combines rapid convergence to interpolation with Critically Damped Momentum (CDM)-based post-interpolation norm minimization, offers a natural solution for selecting low-norm interpolating solutions. Under a local quadratic model of the post-interpolation basin, GROKtimizer provides a quadratic speedup over classical gradient descent, with provable optimality among first-order optimizers. To showcase the applicability of our method, we evaluate GROKtimizer on several synthetic benchmarks common in the classical grokking literature and on various real-world datasets. Finally, we reconcile our findings with the flat-minima hypothesis, highlighting the importance of post-interpolation dynamics in the construction of high-quality, generalizing models.

cs.LG↗

arXiv2026

Integrating Large Language Models into Recommendation via Mutual Augmentation and Adaptive Aggregation

Conventional recommendation methods have achieved notable advancements by harnessing collaborative or sequential information from user behavior. Recently, large language models (LLMs) have gained prominence for their capabilities in understanding and reasoning over textual semantics, and have found utility in various domains, including recommendation. Conventional recommendation methods and LLMs each have their strengths and weaknesses. While conventional methods excel at mining collaborative information and modeling sequential behavior, they struggle with data sparsity and the long-tail problem. LLMs, on the other hand, are proficient at utilizing rich textual contexts but face challenges in mining collaborative or sequential information. Despite their individual successes, there is a significant gap in leveraging their combined potential to enhance recommendation performance. In this paper, we introduce a general and model-agnostic framework known as \textbf{L}arge \textbf{la}nguage model with \textbf{m}utual augmentation and \textbf{a}daptive aggregation for \textbf{Rec}ommendation (\textbf{Llama4Rec}). Llama4Rec synergistically combines conventional and LLM-based recommendation models. Llama4Rec proposes data augmentation and prompt augmentation strategies tailored to enhance the conventional model and LLM respectively. An adaptive aggregation module is adopted to combine the predictions of both kinds of models to refine the final recommendation results. Empirical studies on three real-world datasets validate the superiority of Llama4Rec, demonstrating its consistent outperformance of baseline methods and significant improvements in recommendation performance.

cs.IR↗

arXiv2025

Autoformulation of Mathematical Optimization Models Using LLMs

Mathematical optimization is fundamental to decision-making across diverse domains, from operations research to healthcare. Yet, translating real-world problems into optimization models remains a difficult task, often demanding specialized expertise. This paper approaches the problem of $\textit{autoformulation}$: the automated creation of solver-ready optimization models from natural language problem descriptions. We identify three core challenges of autoformulation: $\textit{(1)}$ the vast, problem-dependent hypothesis space, $\textit{(2)}$ efficient and diverse exploration of this space under uncertainty, and $\textit{(3)}$ evaluation of formulation correctness against problem description. To address these challenges, we present a novel method leveraging $\textit{Large Language Models}$ (LLMs) with $\textit{Monte-Carlo Tree Search}$, exploiting the hierarchical nature of optimization modeling to generate and systematically explore possible formulations. To enhance search efficiency, we introduce symbolic pruning to eliminate trivially equivalent search paths (branches), and employ LLM-based evaluation of partial formulations to guide search. Empirical analysis on linear and mixed-integer programming benchmarks demonstrates our method's effectiveness, with significant performance gains from both LLM-based value estimation and symbolic pruning techniques.

cs.LG↗

arXiv2024

RecRanker: Instruction Tuning Large Language Model as Ranker for Top-k Recommendation

Large Language Models (LLMs) have demonstrated remarkable capabilities and have been extensively deployed across various domains, including recommender systems. Prior research has employed specialized \textit{prompts} to leverage the in-context learning capabilities of LLMs for recommendation purposes. More recent studies have utilized instruction tuning techniques to align LLMs with human preferences, promising more effective recommendations. However, existing methods suffer from several limitations. The full potential of LLMs is not fully elicited due to low-quality tuning data and the overlooked integration of conventional recommender signals. Furthermore, LLMs may generate inconsistent responses for different ranking tasks in the recommendation, potentially leading to unreliable results. In this paper, we introduce \textbf{RecRanker}, tailored for instruction tuning LLMs to serve as the \textbf{Ranker} for top-\textit{k} \textbf{Rec}ommendations. Specifically, we introduce an adaptive sampling module for sampling high-quality, representative, and diverse training data. To enhance the prompt, we introduce a position shifting strategy to mitigate position bias and augment the prompt with auxiliary information from conventional recommendation models, thereby enriching the contextual understanding of the LLM. Subsequently, we utilize the sampled data to assemble an instruction-tuning dataset with the augmented prompts comprising three distinct ranking tasks: pointwise, pairwise, and listwise rankings. We further propose a hybrid ranking method to enhance the model performance by ensembling these ranking tasks. Our empirical evaluations demonstrate the effectiveness of our proposed RecRanker in both direct and sequential recommendation scenarios.

cs.IR↗

arXiv2024

Truncated Non-Uniform Quantization for Distributed SGD

To address the communication bottleneck challenge in distributed learning, our work introduces a novel two-stage quantization strategy designed to enhance the communication efficiency of distributed Stochastic Gradient Descent (SGD). The proposed method initially employs truncation to mitigate the impact of long-tail noise, followed by a non-uniform quantization of the post-truncation gradients based on their statistical characteristics. We provide a comprehensive convergence analysis of the quantized distributed SGD, establishing theoretical guarantees for its performance. Furthermore, by minimizing the convergence error, we derive optimal closed-form solutions for the truncation threshold and non-uniform quantization levels under given communication constraints. Both theoretical insights and extensive experimental evaluations demonstrate that our proposed algorithm outperforms existing quantization schemes, striking a superior balance between communication efficiency and convergence performance.

cs.LG↗

arXiv2024

Improved Quantization Strategies for Managing Heavy-tailed Gradients in Distributed Learning

Gradient compression has surfaced as a key technique to address the challenge of communication efficiency in distributed learning. In distributed deep learning, however, it is observed that gradient distributions are heavy-tailed, with outliers significantly influencing the design of compression strategies. Existing parameter quantization methods experience performance degradation when this heavy-tailed feature is ignored. In this paper, we introduce a novel compression scheme specifically engineered for heavy-tailed gradients, which effectively combines gradient truncation with quantization. This scheme is adeptly implemented within a communication-limited distributed Stochastic Gradient Descent (SGD) framework. We consider a general family of heavy-tail gradients that follow a power-law distribution, we aim to minimize the error resulting from quantization, thereby determining optimal values for two critical parameters: the truncation threshold and the quantization density. We provide a theoretical analysis on the convergence error bound under both uniform and non-uniform quantization scenarios. Comparative experiments with other benchmarks demonstrate the effectiveness of our proposed method in managing the heavy-tailed gradients in a distributed learning environment.

cs.LG↗

arXiv2023

Adaptive Top-K in SGD for Communication-Efficient Distributed Learning

Distributed stochastic gradient descent (SGD) with gradient compression has become a popular communication-efficient solution for accelerating distributed learning. One commonly used method for gradient compression is Top-K sparsification, which sparsifies the gradients by a fixed degree during model training. However, there has been a lack of an adaptive approach to adjust the sparsification degree to maximize the potential of the model's performance or training speed. This paper proposes a novel adaptive Top-K in SGD framework that enables an adaptive degree of sparsification for each gradient descent step to optimize the convergence performance by balancing the trade-off between communication cost and convergence error. Firstly, an upper bound of convergence error is derived for the adaptive sparsification scheme and the loss function. Secondly, an algorithm is designed to minimize the convergence error under the communication cost constraints. Finally, numerical results on the MNIST and CIFAR-10 datasets demonstrate that the proposed adaptive Top-K algorithm in SGD achieves a significantly better convergence rate compared to state-of-the-art methods, even after considering error compensation.

cs.LG↗

arXiv2023

Unsupervised Massive MIMO Channel Estimation with Dual-Path Knowledge-Aware Auto-Encoders

In this paper, an unsupervised deep learning framework based on dual-path model-driven variational auto-encoders (VAE) is proposed for angle-of-arrivals (AoAs) and channel estimation in massive MIMO systems. Specifically designed for channel estimation, the proposed VAE differs from the original VAE in two aspects. First, the encoder is a dual-path neural network, where one path uses the received signal to estimate the path gains and path angles, and another uses the correlation matrix of the received signal to estimate AoAs. Second, the decoder has fixed weights that implement the signal propagation model, instead of learnable parameters. This knowledge-aware decoder forces the encoder to output meaningful physical parameters of interests (i.e., path gains, path angles, and AoAs), which cannot be achieved by original VAE. Rigorous analysis is carried out to characterize the multiple global optima and local optima of the estimation problem, which motivates the design of the dual-path encoder. By alternating between the estimation of path gains, path angles and the estimation of AoAs, the encoder is proved to converge. To further improve the convergence performance, a low-complexity procedure is proposed to find good initial points. Numerical results validate theoretical analysis and demonstrate the performance improvements of our proposed framework.

eess.SP↗

arXiv2023

PerFedRec++: Enhancing Personalized Federated Recommendation with Self-Supervised Pre-Training

Federated recommendation systems employ federated learning techniques to safeguard user privacy by transmitting model parameters instead of raw user data between user devices and the central server. Nevertheless, the current federated recommender system faces challenges such as heterogeneity and personalization, model performance degradation, and communication bottleneck. Previous studies have attempted to address these issues, but none have been able to solve them simultaneously. In this paper, we propose a novel framework, named PerFedRec++, to enhance the personalized federated recommendation with self-supervised pre-training. Specifically, we utilize the privacy-preserving mechanism of federated recommender systems to generate two augmented graph views, which are used as contrastive tasks in self-supervised graph learning to pre-train the model. Pre-training enhances the performance of federated models by improving the uniformity of representation learning. Also, by providing a better initial state for federated training, pre-training makes the overall training converge faster, thus alleviating the heavy communication burden. We then construct a collaborative graph to learn the client representation through a federated graph neural network. Based on these learned representations, we cluster users into different user groups and learn personalized models for each cluster. Each user learns a personalized model by combining the global federated model, the cluster-level federated model, and its own fine-tuned local model. Experiments on three real-world datasets show that our proposed method achieves superior performance over existing methods.

cs.IR↗

arXiv2022

Towards Communication Efficient and Fair Federated Personalized Sequential Recommendation

Federated recommendations leverage the federated learning (FL) techniques to make privacy-preserving recommendations. Though recent success in the federated recommender system, several vital challenges remain to be addressed: (i) The majority of federated recommendation models only consider the model performance and the privacy-preserving ability, while ignoring the optimization of the communication process; (ii) Most of the federated recommenders are designed for heterogeneous systems, causing unfairness problems during the federation process; (iii) The personalization techniques have been less explored in many federated recommender systems. In this paper, we propose a Communication efficient and Fair personalized Federated personalized Sequential Recommendation algorithm (CF-FedSR) to tackle these challenges. CF-FedSR introduces a communication-efficient scheme that employs adaptive client selection and clustering-based sampling to accelerate the training process. A fairness-aware model aggregation algorithm that can adaptively capture the data and performance imbalance among different clients to address the unfairness problems is proposed. The personalization module assists clients in making personalized recommendations and boosts the recommendation performance via local fine-tuning and model adaption. Extensive experimental results show the effectiveness and efficiency of our proposed method.

cs.IR↗

arXiv2022

HySAGE: A Hybrid Static and Adaptive Graph Embedding Network for Context-Drifting Recommendations

The recent popularity of edge devices and Artificial Intelligent of Things (AIoT) has driven a new wave of contextual recommendations, such as location based Point of Interest (PoI) recommendations and computing resource-aware mobile app recommendations. In many such recommendation scenarios, contexts are drifting over time. For example, in a mobile game recommendation, contextual features like locations, battery, and storage levels of mobile devices are frequently drifting over time. However, most existing graph-based collaborative filtering methods are designed under the assumption of static features. Therefore, they would require frequent retraining and/or yield graphical models burgeoning in sizes, impeding their suitability for context-drifting recommendations. In this work, we propose a specifically tailor-made Hybrid Static and Adaptive Graph Embedding (HySAGE) network for context-drifting recommendations. Our key idea is to disentangle the relatively static user-item interaction and rapidly drifting contextual features. Specifically, our proposed HySAGE network learns a relatively static graph embedding from user-item interaction and an adaptive embedding from drifting contextual features. These embeddings are incorporated into an interest network to generate the user interest in some certain context. We adopt an interactive attention module to learn the interactions among static graph embeddings, adaptive contextual embeddings, and user interest, helping to achieve a better final representation. Extensive experiments on real-world datasets demonstrate that HySAGE significantly improves the performance of the existing state-of-the-art recommendation algorithms.

cs.IR↗

arXiv2022

Personalized Federated Recommendation via Joint Representation Learning, User Clustering, and Model Adaptation

Federated recommendation applies federated learning techniques in recommendation systems to help protect user privacy by exchanging models instead of raw user data between user devices and the central server. Due to the heterogeneity in user's attributes and local data, attaining personalized models is critical to help improve the federated recommendation performance. In this paper, we propose a Graph Neural Network based Personalized Federated Recommendation (PerFedRec) framework via joint representation learning, user clustering, and model adaptation. Specifically, we construct a collaborative graph and incorporate attribute information to jointly learn the representation through a federated GNN. Based on these learned representations, we cluster users into different user groups and learn personalized models for each cluster. Then each user learns a personalized model by combining the global federated model, the cluster-level federated model, and the user's fine-tuned local model. To alleviate the heavy communication burden, we intelligently select a few representative users (instead of randomly picked users) from each cluster to participate in training. Experiments on real-world datasets show that our proposed method achieves superior performance over existing methods.

cs.IR↗

arXiv2019

Model Predictive Control-Based Battery Scheduling and Incentives to Manipulate Demand Response Baselines

We study operations of a battery energy storage system under a baseline-based demand response (DR) program with an uncertain schedule of DR events. Baseline-based DR programs may provide undesired incentives to inflate baseline consumption in non-event days, in order to increase "apparent" DR reduction in event days and secure higher DR payments. Our goal is to identify and quantify such incentives. To understand customer decisions, we formulate the problem of determining hourly battery charging and discharge schedules to minimize expected net costs, defined as energy purchase costs minus energy export rebates and DR payments, over a sufficiently long time horizon (e.g., a year). The complexity of this stochastic optimization problem grows exponentially with the time horizon considered. To obtain computationally tractable solutions, we propose using multistage model predictive control with scenario sampling. Numerical results indicate that our solutions are near optimal (e.g., within 3% from the optimum in the test cases). Finally, we apply our solutions to study an example residential customer with solar photovoltaic and battery systems participating in a typical existing baseline-based DR program. Results reveal that over 66% of the average apparent load reduction during DR events could result from inflation of baseline consumption during non-event days.

eess.SY↗

arXiv2019

Competitive Online Optimization under Inventory Constraints

This paper studies online optimization under inventory (budget) constraints. While online optimization is a well-studied topic, versions with inventory constraints have proven difficult. We consider a formulation of inventory-constrained optimization that is a generalization of the classic one-way trading problem and has a wide range of applications. We present a new algorithmic framework, \textsf{CR-Pursuit}, and prove that it achieves the minimal competitive ratio among all deterministic algorithms (up to a problem-dependent constant factor) for inventory-constrained online optimization. Our algorithm and its analysis not only simplify and unify the state-of-the-art results for the standard one-way trading problem, but they also establish novel bounds for generalizations including concave revenue functions. For example, for one-way trading with price elasticity, the \textsf{CR-Pursuit} algorithm achieves a competitive ratio that is within a small additive constant (i.e., 1/3) to the lower bound of $\ln θ+1$, where $θ$ is the ratio between the maximum and minimum base prices.

cs.PF↗

arXiv2018

A Dynamic Jamming Game for Real-Time Status Updates

We study timely status updates of a real-time system in an adversarial setting. The system samples a physical process, and sends the samples from the source (e.g., a sensor) to the destination (e.g, a control center) through a channel. For real-time monitoring/control tasks, it is crucial for the system to update the status of the physical process "timely". We measure the timeliness of status updates by the time elapsed since the latest update at the destination was generated at the source, and define the time elapsed as age of information, or age in short. To sabotage the system, an attacker aims to maximize the age by jamming the channel and hence causing delay in status updates. The system aims to minimize the age by judiciously choosing when to sample and send the updates. We model the ongoing repeated interaction between the attacker and the system as a dynamic game. In each stage game, the attacker chooses the jamming time according to the jamming time distribution, and the system responds by delaying the sampling according to the sampling policy. We prove that there exists a unique stationary equilibrium in the game, and provide a complete analytical characterization of the equilibrium. Our results shed lights on how the attacker sabotages the system and how the system should defend against the attacker.

cs.IT↗

arXiv2015

Efficient Interference Management Policies for Femtocell Networks

Managing interference in a network of macrocells underlaid with femtocells presents an important, yet challenging problem. A majority of spatial (frequency/time) reuse based approaches partition the users based on coloring the interference graph, which is shown to be suboptimal. Some spatial time reuse based approaches schedule the maximal independent sets (MISs) in a cyclic, (weighted) round-robin fashion, which is inefficient for delay-sensitive applications. Our proposed policies schedule the MISs in a non-cyclic fashion, which aim to optimize any given network performance criterion for delay-sensitive applications while fulfilling minimum throughput requirements of the users. Importantly, we do not take the interference graph as given as in existing works; we propose an optimal construction of the interference graph. We prove that under certain conditions, the proposed policy achieves the optimal network performance. For large networks, we propose a low-complexity algorithm for computing the proposed policy. We show that the policy computed achieves a constant competitive ratio (with respect to the optimal network performance), which is independent of the network size, under wide range of deployment scenarios. The policy can be implemented in a decentralized manner by the users. Compared to the existing policies, our proposed policies can achieve improvement of up to 130 % in large-scale deployments.

cs.IT↗

arXiv2015

Distributed Interference Management Policies for Heterogeneous Small Cell Networks

We study the problem of interference management in large-scale small cell networks, where each user equipment (UE) needs to determine in a distributed manner when and at what power level it should transmit to its serving small cell base station (SBS) such that a given network performance criterion is maximized subject to minimum quality of service (QoS) requirements by the UEs. We first propose a distributed algorithm for the UE-SBS pairs to find a subset of weakly interfering UE-SBS pairs, namely the maximal independent sets (MISs) of the interference graph in logarithmic time (with respect to the number of UEs). Then we propose a novel problem formulation which enables UE-SBS pairs to determine the optimal fractions of time occupied by each MIS in a distributed manner. We analytically bound the performance of our distributed policy in terms of the competitive ratio with respect to the optimal network performance, which is obtained in a centralized manner with NP (non-deterministic polynomial time) complexity. Remarkably, the competitive ratio is independent of the network size, which guarantees scalability in terms of performance for arbitrarily large networks. Through simulations, we show that our proposed policies achieve significant performance improvements (from 150% to 700%) over the existing policies.

cs.IT↗

arXiv2014

Incentive Design in Peer Review: Rating and Repeated Endogenous Matching

Peer review (e.g., grading assignments in Massive Open Online Courses (MOOCs), academic paper review) is an effective and scalable method to evaluate the products (e.g., assignments, papers) of a large number of agents when the number of dedicated reviewing experts (e.g., teaching assistants, editors) is limited. Peer review poses two key challenges: 1) identifying the reviewers' intrinsic capabilities (i.e., adverse selection) and 2) incentivizing the reviewers to exert high effort (i.e., moral hazard). Some works in mechanism design address pure adverse selection using one-shot matching rules, and pure moral hazard was addressed in repeated games with exogenously given and fixed matching rules. However, in peer review systems exhibiting both adverse selection and moral hazard, one-shot or exogenous matching rules do not link agents' current behavior with future matches and future payoffs, and as we prove, will induce myopic behavior (i.e., exerting the lowest effort) resulting in the lowest review quality. In this paper, we propose for the first time a solution that simultaneously solves adverse selection and moral hazard. Our solution exploits the repeated interactions of agents, utilizes ratings to summarize agents' past review quality, and designs matching rules that endogenously depend on agents' ratings. Our proposed matching rules are easy to implement and require no knowledge about agents' private information (e.g., their benefit and cost functions). Yet, they are effective in guiding the system to an equilibrium where the agents are incentivized to exert high effort and receive ratings that precisely reflect their review quality. Using several illustrative examples, we quantify the significant performance gains obtained by our proposed mechanism as compared to existing one-shot or exogenous matching rules.

cs.GT↗