Source author record

Hongzhe Liu

Hongzhe Liu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Computer Vision Cryptography and Security Genomics Hardware Architecture Machine Learning math.OC

Catalog footprint

What is connected

4works

6topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

Distributed Algorithm Over Time-Varying Unbalanced Topologies for Optimization Problem Subject to Multiple Local Constraints

This paper studies the distributed optimization problem with possibly nonidentical local constraints, where its global objective function is composed of $N$ convex functions. The aim is to solve the considered optimization problem in a distributed manner over time-varying unbalanced directed topologies by using only local information and performing only local computations. Towards this end, a new distributed discrete-time algorithm is developed by synthesizing the row stochastic matrices sequence and column stochastic matrices sequence analysis technique. Furthermore, for the developed distributed discrete-time algorithm, its convergence property to the optimal solution as well as its convergence rate are established under some mild assumptions. Numerical simulations are finally presented to verify the theoretical results.

preprint2022arXiv

Toward Minimal Misalignment at Minimal Cost in One-Stage and Anchor-Free Object Detection

Common object detection models consist of classification and regression branches, due to different task drivers, these two branches have different sensibility to the features from the same scale level and the same spatial location. The point-based prediction method, which is based on the assumption that the high classification confidence point has the high regression quality, leads to the misalignment problem. Our analysis shows, the problem is further composed of scale misalignment and spatial misalignment specifically. We aim to resolve the phenomenon at minimal cost: a minor adjustment of the head network and a new label assignment method replacing the rigid one. Our experiments show that, compared to the baseline FCOS, a one-stage and anchor-free object detection model, our model consistently get around 3 AP improvement with different backbones, demonstrating both simplicity and efficiency of our method.

preprint2016arXiv

Tardis 2.0: Optimized Time Traveling Coherence for Relaxed Consistency Models

Cache coherence scalability is a big challenge in shared memory systems. Traditional protocols do not scale due to the storage and traffic overhead of cache invalidation. Tardis, a recently proposed coherence protocol, removes cache invalidation using logical timestamps and achieves excellent scalability. The original Tardis protocol, however, only supports the Sequential Consistency (SC) memory model, limiting its applicability. Tardis also incurs extra network traffic on some benchmarks due to renew messages, and has suboptimal performance when the program uses spinning to communicate between threads. In this paper, we address these downsides of Tardis protocol and make it significantly more practical. Specifically, we discuss the architectural, memory system and protocol changes required in order to implement the TSO consistency model on Tardis, and prove that the modified protocol satisfies TSO. We also describe modifications for Partial Store Order (PSO) and Release Consistency (RC). Finally, we propose optimizations for better leasing policies and to handle program spinning. On a set of benchmarks, optimized Tardis improves on a full-map directory protocol in the metrics of performance, storage and network traffic, while being simpler to implement.

preprint2015arXiv

Supporting Regularized Logistic Regression Privately and Efficiently

As one of the most popular statistical and machine learning models, logistic regression with regularization has found wide adoption in biomedicine, social sciences, information technology, and so on. These domains often involve data of human subjects that are contingent upon strict privacy regulations. Increasing concerns over data privacy make it more and more difficult to coordinate and conduct large-scale collaborative studies, which typically rely on cross-institution data sharing and joint analysis. Our work here focuses on safeguarding regularized logistic regression, a widely-used machine learning model in various disciplines while at the same time has not been investigated from a data security and privacy perspective. We consider a common use scenario of multi-institution collaborative studies, such as in the form of research consortia or networks as widely seen in genetics, epidemiology, social sciences, etc. To make our privacy-enhancing solution practical, we demonstrate a non-conventional and computationally efficient method leveraging distributing computing and strong cryptography to provide comprehensive protection over individual-level and summary data. Extensive empirical evaluation on several studies validated the privacy guarantees, efficiency and scalability of our proposal. We also discuss the practical implications of our solution for large-scale studies and applications from various disciplines, including genetic and biomedical studies, smart grid, network analysis, etc.