Searcharxiv⌕ Search

arXiv subjects

Hong Guan

Publications and source records attributed to Hong Guan.

13 recordsLinked to original sources
arXiv2026

CACTUSDB: Unlock Co-Optimization Opportunities for SQL and AI/ML Inferences

There is a growing demand for supporting inference queries that combine Structured Query Language (SQL) and Artificial Intelligence / Machine Learning (AI/ML) model inferences in database systems, to avoid data denormalization and transfer, facilitate management, and alleviate privacy concerns. Co-optimization techniques for executing inference queries in database systems without accuracy loss fall into four categories: (O1) Relational algebra optimization treating AI/ML models as black-box user-defined functions (UDFs); (O2) Factorized AI/ML inferences; (O3) Tensor-relational transformation; and (O4) General cross-optimization techniques. However, we found none of the existing database systems support all these techniques simultaneously, resulting in suboptimal performance. In this work, we identify two key challenges to address the above problem: (1) the difficulty of unifying all co-optimization techniques that involve disparate data and computation abstractions in one system; and (2) the lack of an optimizer that can effectively explore the exponential search space. To address these challenges, we present CactusDB, a novel system built atop Velox - a high-performance, UDF-centric database engine, open-sourced by Meta. CactusDB features a three-level Intermediate Representations (IR) that supports relational operators, expression operators, and ML functions to enable flexible optimization of arbitrary sub-computations. Additionally, we propose a novel Monte-Carlo Tree Search (MCTS)-based optimizer with query embedding, co-designed with our unique three-level IR, enabling shared and reusable optimization knowledge across different queries. Evaluation of 12 representative inference workloads and 2,000 randomly generated inference queries on well-known datasets, such as MovieLens and TPCx-AI, shows that CactusDB achieves up to 441 times speedup compared to alternative systems.

cs.DB↗
arXiv2025

FedSGT: Exact Federated Unlearning via Sequential Group-based Training

Federated Learning (FL) enables collaborative, privacy-preserving model training, but supporting the "Right to be Forgotten" is especially challenging because data influences the model through distributed and interleaved client updates. Existing exact unlearning methods typically require frequent retraining from scratch, resulting in high communication cost and long service downtime. To address this, we propose Federated Sequential Group-based Training (FedSGT), an exact unlearning framework for FL. FedSGT partitions the data into uniform groups, and each client may participate in multiple groups. To control communication overhead, each client can limit the number of groups it contributes to. FedSGT then trains multiple sequences of Parameter-Efficient Fine-Tuning (PEFT) modules, each corresponding to a different group permutation. Since the PEFT modules are lightweight and maintained server-side, FedSGT isolates the influence of different data groups into independent modules without incurring significant storage overhead and communication cost. Exact unlearning is thus achieved instantly by deactivating the modules corresponding to the group containing the unlearned data. Furthermore, using multiple training sequences helps maintain high model utility as deletion requests accumulate. We provide a rigorous theoretical analysis of both the deletion rate -- expected number of deletions before retraining is needed -- and the expected model performance. Experiments on various tasks demonstrate that FedSGT achieves a significantly longer service maintenance under multiple unlearning requests while maintaining comparable learning performance and training efficiency to other exact unlearning baselines. Extensive ablation studies validate the robustness of our method across a wide range of parameter settings.

cs.CR↗
arXiv2025

Privacy and Accuracy-Aware AI/ML Model Deduplication

With the growing adoption of privacy-preserving machine learning algorithms, such as Differentially Private Stochastic Gradient Descent (DP-SGD), training or fine-tuning models on private datasets has become increasingly prevalent. This shift has led to the need for models offering varying privacy guarantees and utility levels to satisfy diverse user requirements. However, managing numerous versions of large models introduces significant operational challenges, including increased inference latency, higher resource consumption, and elevated costs. Model deduplication is a technique widely used by many model serving and database systems to support high-performance and low-cost inference queries and model diagnosis queries. However, none of the existing model deduplication works has considered privacy, leading to unbounded aggregation of privacy costs for certain deduplicated models and inefficiencies when applied to deduplicate DP-trained models. We formalize the problems of deduplicating DP-trained models for the first time and propose a novel privacy- and accuracy-aware deduplication mechanism to address the problems. We developed a greedy strategy to select and assign base models to target models to minimize storage and privacy costs. When deduplicating a target model, we dynamically schedule accuracy validations and apply the Sparse Vector Technique to reduce the privacy costs associated with private validation data. Compared to baselines that do not provide privacy guarantees, our approach improved the compression ratio by up to $35\times$ for individual models (including large language models and vision transformers). We also observed up to $43\times$ inference speedup due to the reduction of I/O operations.

cs.CR↗
arXiv2024

IDNet: A Novel Dataset for Identity Document Analysis and Fraud Detection

Effective fraud detection and analysis of government-issued identity documents, such as passports, driver's licenses, and identity cards, are essential in thwarting identity theft and bolstering security on online platforms. The training of accurate fraud detection and analysis tools depends on the availability of extensive identity document datasets. However, current publicly available benchmark datasets for identity document analysis, including MIDV-500, MIDV-2020, and FMIDV, fall short in several respects: they offer a limited number of samples, cover insufficient varieties of fraud patterns, and seldom include alterations in critical personal identifying fields like portrait images, limiting their utility in training models capable of detecting realistic frauds while preserving privacy. In response to these shortcomings, our research introduces a new benchmark dataset, IDNet, designed to advance privacy-preserving fraud detection efforts. The IDNet dataset comprises 837,060 images of synthetically generated identity documents, totaling approximately 490 gigabytes, categorized into 20 types from $10$ U.S. states and 10 European countries. We evaluate the utility and present use cases of the dataset, illustrating how it can aid in training privacy-preserving fraud detection methods, facilitating the generation of camera and video capturing of identity documents, and testing schema unification and other identity document management functionalities.

cs.CV↗
arXiv2024

Declarative Privacy-Preserving Inference Queries

Detecting inference queries running over personal attributes and protecting such queries from leaking individual information requires tremendous effort from practitioners. To tackle this problem, we propose an end-to-end workflow for automating privacy-preserving inference queries including the detection of subqueries that involve AI/ML model inferences on sensitive attributes. Our proposed novel declarative privacy-preserving workflow allows users to specify "what private information to protect" rather than "how to protect". Under the hood, the system automatically chooses privacy-preserving plans and hyper-parameters.

cs.DB↗
arXiv2023

Automatic Data Transformation Using Large Language Model: An Experimental Study on Building Energy Data

Existing approaches to automatic data transformation are insufficient to meet the requirements in many real-world scenarios, such as the building sector. First, there is no convenient interface for domain experts to provide domain knowledge easily. Second, they require significant training data collection overheads. Third, the accuracy suffers from complicated schema changes. To bridge this gap, we present a novel approach that leverages the unique capabilities of large language models (LLMs) in coding, complex reasoning, and zero-shot learning to generate SQL code that transforms the source datasets into the target datasets. We demonstrate the viability of this approach by designing an LLM-based framework, termed SQLMorpher, which comprises a prompt generator that integrates the initial prompt with optional domain knowledge and historical patterns in external databases. It also implements an iterative prompt optimization mechanism that automatically improves the prompt based on flaw detection. The key contributions of this work include (1) pioneering an end-to-end LLM-based solution for data transformation, (2) developing a benchmark dataset of 105 real-world building energy data transformation problems, and (3) conducting an extensive empirical evaluation where our approach achieved 96% accuracy in all 105 problems. SQLMorpher demonstrates the effectiveness of utilizing LLMs in complex, domain-specific challenges, highlighting the potential of their potential to drive sustainable solutions.

cs.DB↗
arXiv2023

A Comparison of Decision Forest Inference Platforms from A Database Perspective

Decision forest, including RandomForest, XGBoost, and LightGBM, is one of the most popular machine learning techniques used in many industrial scenarios, such as credit card fraud detection, ranking, and business intelligence. Because the inference process is usually performance-critical, a number of frameworks were developed and dedicated for decision forest inference, such as ONNX, TreeLite from Amazon, TensorFlow Decision Forest from Google, HummingBird from Microsoft, Nvidia FIL, and lleaves. However, these frameworks are all decoupled with data management frameworks. It is unclear whether in-database inference will improve the overall performance. In addition, these frameworks used different algorithms, optimization techniques, and parallelism models. It is unclear how these implementations will affect the overall performance and how to make design decisions for an in-database inference framework. In this work, we investigated the above questions by comprehensively comparing the end-to-end performance of the aforementioned inference frameworks and netsDB, an in-database inference framework we implemented. Through this study, we identified that netsDB is best suited for handling small-scale models on large-scale datasets and all-scale models on small-scale datasets, for which it achieved up to hundreds of times of speedup. In addition, the relation-centric representation we proposed significantly improved netsDB's performance in handling large-scale models, while the model reuse optimization we proposed further improved netsDB's performance in handling small-scale datasets.

cs.DB↗
arXiv2021

A Bayesian Approach for Medical Inquiry and Disease Inference in Automated Differential Diagnosis

We propose a Bayesian approach for both medical inquiry and disease inference, the two major phases in differential diagnosis. Unlike previous work that simulates data from given probabilities and uses ML algorithms on them, we directly use the Quick Medical Reference (QMR) belief network, and apply Bayesian inference in the inference phase and Bayesian experimental design in the inquiry phase. Moreover, we improve the inquiry phase by extending the Bayesian experimental design framework from one-step search to multi-step search. Our approach has some practical advantages as it is interpretable, free of costly training, and able to adapt to new changes without any additional effort. Our experiments show that our approach achieves new state-of-the-art results on two simulated datasets, SymCAT and HPO, and competitive results on two diagnosis dialogue datasets, Muzhi and Dxy.

cs.AI↗
arXiv2021

COVID-19: Comparative Analysis of Methods for Identifying Articles Related to Therapeutics and Vaccines without Using Labeled Data

Here we proposed an approach to analyze text classification methods based on the presence or absence of task-specific terms (and their synonyms) in the text. We applied this approach to study six different transfer-learning and unsupervised methods for screening articles relevant to COVID-19 vaccines and therapeutics. The analysis revealed that while a BERT model trained on search-engine results generally performed well, it miss-classified relevant abstracts that did not contain task-specific terms. We used this insight to create a more effective unsupervised ensemble.

cs.IR↗
arXiv2020

Robustly Pre-trained Neural Model for Direct Temporal Relation Extraction

Background: Identifying relationships between clinical events and temporal expressions is a key challenge in meaningfully analyzing clinical text for use in advanced AI applications. While previous studies exist, the state-of-the-art performance has significant room for improvement. Methods: We studied several variants of BERT (Bidirectional Encoder Representations using Transformers) some involving clinical domain customization and the others involving improved architecture and/or training strategies. We evaluated these methods using a direct temporal relations dataset which is a semantically focused subset of the 2012 i2b2 temporal relations challenge dataset. Results: Our results show that RoBERTa, which employs better pre-training strategies including using 10x larger corpus, has improved overall F measure by 0.0864 absolute score (on the 1.00 scale) and thus reducing the error rate by 24% relative to the previous state-of-the-art performance achieved with an SVM (support vector machine) model. Conclusion: Modern contextual language modeling neural networks, pre-trained on a large corpus, achieve impressive performance even on highly-nuanced clinical temporal relation tasks.

cs.CL↗
arXiv2014

Fourth-order gravity gradient torque of spacecraft orbiting asteroids

The dynamical behavior of spacecraft around asteroids is a key element in design of such missions. An asteroid's irregular shape, non-spherical mass distribution and its rotational sate make the dynamics of spacecraft quite complex. This paper focuses on the gravity gradient torque of spacecraft around non-spherical asteroids. The gravity field of the asteroid is approximated as a 2nd degree and order-gravity field with harmonic coefficients C20 and C22. By introducing the spacecraft's higher-order inertia integrals, a full fourth-order gravity gradient torque model of the spacecraft is established through the gravitational potential derivatives. Our full fourth-order model is more precise than previous fourth-order model due to the consideration of higher-order inertia integrals of the spacecraft. Some interesting conclusions about the gravity gradient torque model are reached. Then a numerical simulation is carried out to verify our model. In the numerical simulation, a special spacecraft consisted of 36 point masses connected by rigid massless rods is considered. We assume that the asteroid is in a uniform rotation around its maximum-moment principal axis, and the spacecraft is on the stationary orbit in the equatorial plane. Simulation results show that the motion of previous fourth-order model is quite different from the exact motion, while our full fourth-order model fits the exact motion very well. And our model is precise enough for practical applications.

astro-ph.EP↗
arXiv1999

Einstein-Podolsky-Rosen Paradox and Antiparticle

The original version of Einstein-Podolsky-Rosen (EPR) paradox is discussed to show the completeness of Quantum Mechanics (QM). The unique solution leads to the wave function of antiparticle unambiguously, which implies the essential conformity between QM and Special Relativity (SR).

quant-ph↗