L
Lifan Xu
Researcher at University of Delaware
Publications - 11
Citations - 939
Lifan Xu is an academic researcher from University of Delaware. The author has contributed to research in topics: Graph kernel & Deep learning. The author has an hindex of 9, co-authored 11 publications receiving 768 citations. Previous affiliations of Lifan Xu include University UCINF.
Papers
More filters
Proceedings ArticleDOI
Auto-tuning a high-level language targeted to GPU codes
TL;DR: This work performs auto-tuning on a large optimization space on GPU kernels, focusing on loop permutation, loop unrolling, tiling, and specifying which loop(s) to parallelize, and shows results on convolution kernels, codes in the PolyBench suite, and an implementation of belief propagation for stereo vision.
Proceedings ArticleDOI
TOP-PIM: throughput-oriented programmable processing in memory
Dong Ping Zhang,Nuwan Jayasena,Alexander Lyashevsky,Joseph L. Greathouse,Lifan Xu,Michael Ignatowski +5 more
TL;DR: This work explores the use of 3D die stacking to move memory-intensive computations closer to memory and introduces a methodology for rapid design space exploration by analytically predicting performance and energy of in-memory processors based on metrics obtained from execution on today's GPU hardware.
Book ChapterDOI
HADM: Hybrid Analysis for Detection of Malware
TL;DR: The Android operating system is now the platform most targeted by malware, creating an urgent need for effective defense mechanisms to protect Android-enabled devices.
Journal ArticleDOI
Hierarchical fractional-step approximations and parallel kinetic Monte Carlo algorithms
TL;DR: A spatial domain decomposition of the Markov operator (generator) that describes the evolution of all observables according to the kinetic Monte Carlo algorithm is developed and formed, which can be tailored to specific hierarchical parallel architectures such as multi-core processors or clusters of Graphical Processing Units (GPUs).
Proceedings ArticleDOI
Dynamic Android Malware Classification Using Graph-Based Representations
TL;DR: This paper implements three traditional feature-vector-based representations for Android system calls and proposes a novel graph-based representation that is able to improve the classification accuracy over the corresponding feature- vector-basedrepresentations from the same input.