A digital 3D TCAM accelerator for the inference phase of Random Forest
DescriptionRandom forest is a popular ensemble machine-learning method for classification and regression tasks. However, the irregular tree shapes and non-deterministic memory access patterns make it hard for the current von Neumann architecture to handle it efficiently. This paper proposes a digital 3D TCAM-based accelerator for the random forest, adopting the idea of processing-in-memory to reduce data movement. The proposed method can provide real-time inference with low energy consumption, making it suitable for edge or embedded environments. In the experiment, the proposed accelerator achieves an average of 3.13 times higher throughput with 22 times more energy saving than the GPU approach.
Event Type
Research Manuscript
TimeTuesday, July 11th1:55pm - 2:10pm PDT
Location3010, 3rd Floor
In-memory and Near-memory Computing Architectures, Applications and Systems