Deploying DNNs in P4 switches seemed impossible—until we distilled them into cascaded LUTs. The result? Lossless accuracy, wirespeed performance, and 96% less memory than traditional methods. In-network AI just got real. 🌟 Follow us on LinkedIn! Check the updates…
No problem. Our cascaded LUT distillation maps DNNs into P4 flow tables, enabling wirespeed inference without hardware changes. Tested on DDoS detection, it achieves 93% F1-score with 8-bit inputs. 🌟 Follow us on LinkedIn! Check the updates from the…
P4 switches weren’t built for DNNs—but our LUT distillation changes that. By mapping neurons to flow tables, we enable complex AI models in hardware-constrained pipelines, with no accuracy loss. Tested on DDoS detection, it’s faster and leaner than GPU offloading.…
Detecting DDoS attacks at wirespeed requires in-network AI. Our P4 LUT distillation deploys a DNN-based mitigator without ALUs, achieving 93% F1-score on UNSW-NB15 dataset. The secret? Cascaded lookup tables that replace neurons with flow matches. 🌟 Follow us on LinkedIn!…
Traditional DNN quantization sacrifices accuracy—our LUT-based distillation doesn’t. By mapping neurons to flow tables, we deploy complex DNNs in P4 switches with zero information loss. Demonstrated on UNSW-NB15 dataset, it hits 93% F1-score for DDoS detection. 🌟 Follow us on…
Hardware P4 switches lack ALUs and fast stateful memory, making DNN deployment impossible—until now. Our cascaded LUT distillation maps any DNN into lookup tables, enabling wirespeed inference with no performance loss. Tested on DDoS mitigation, it achieves 93% F1-score with…
