CYCLE
CYCLE
News
People
Events
Publications
Contact
Jie Zhao
Latest
Scalable Detection of Floating-point Errors via Adaptive Parallel Subdomain Search
Optimizing Deep Learning Inference Efficiency through Block Dependency Analysis
Magneto: Accelerating Parallel Structures in DNNs via Co-Optimization of Operators (poster)
Post-Link Outlining for Code Size Reduction
Arfa: an Agile Regime-based Floating-point Optimization Approach for Rounding Errors
Enabling Tensor Language Model to Assist in Generating High-Performance Tensor Programs for Deep Learning
A Holistic Approach to Automatic Mixed-Precision Code Generation and Tuning for Affine Programs
Modeling the Interplay between Loop Tiling and Fusion in Optimizing Compilers Using Affine Relations
Apollo: Automatic Partition-based Operator Fusion through Layer by Layer Optimization
Eiffel: Inferring Input Ranges of Significant Floating-point Errors via Polynomial Extrapolation
SIRIUS: Harvesting Whole-Program Optimization Opportunities for DNNs
Effectively Scheduling Computational Graphs of Deep Neural Networks toward Their Domain-Specific Accelerators
Automatically Generating High-performance Matrix Multiplication Kernels on the Latest Sunway Processor
Parallelizing Neural Network Models Effectively on GPU by Implementing Reductions Atomically
AKG: automatic kernel generation for neural processing units using polyhedral transformations
Optimizing the Memory Hierarchy by Compositing Automatic Transformations on Computations and Data
Flextended Tiles: A Flexible Extension of Overlapped Tiles for Polyhedral Compilation
A polyhedral compilation framework for loops with dynamic data-dependent bounds
Cite
×