HP INCFollow


As more AI modes are adopted in multiple industries, the security of AI becomes one of the top priorities in business. The failure in AI model protection puts damage to business in terms of user experience, privacy and operation. Model attack can start at any point from model training to model inference. Although many studies have presented several solutions, there are limitations in terms of prediction performance, latency and robustness to unseen attack.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Share Alike 4.0 License.