Exciting News! TreeHouse Accepted for Publication in ACM-TECS

We’re thrilled to announce that our work, TreeHouse: An MLIR-based Compilation Flow for Real-Time Tree-Based Inference, has been accepted for publication in the prestigious ACM Transactions on Embedded Computing Systems (TECS)!

Tree-based ensemble models, such as random forests and boosting trees, have become essential for machine learning applications, especially in resource-constrained environments like embedded systems. However, achieving real-time performance for these models on edge devices is challenging due to their intricate branching and reliance on floating-point operations. With TreeHouse, we introduce a powerful solution that leverages the MLIR framework to streamline and optimize the compilation flow for tree-based models in real-time inference settings.

Read the full paper here and dive into the technical breakthroughs that make TreeHouse a cornerstone for your tree-based inference!

*the post is mostly generated via ChatGPT4o