Xenofon Koutsoukos

Open Set Recognition using Vision Transformer with an Additional Detection Head

Mar 16, 2022
Feiyang Cai, Zhenkai Zhang, Jie Liu, Xenofon Koutsoukos

Figure 2 for Open Set Recognition using Vision Transformer with an Additional Detection Head

Figure 3 for Open Set Recognition using Vision Transformer with an Additional Detection Head

Figure 4 for Open Set Recognition using Vision Transformer with an Additional Detection Head

* under review

Access Paper or Ask Questions

Reliable Probability Intervals For Classification Using Inductive Venn Predictors Based on Distance Learning

Oct 07, 2021
Dimitrios Boursinos, Xenofon Koutsoukos

Figure 2 for Reliable Probability Intervals For Classification Using Inductive Venn Predictors Based on Distance Learning

Figure 3 for Reliable Probability Intervals For Classification Using Inductive Venn Predictors Based on Distance Learning

Figure 4 for Reliable Probability Intervals For Classification Using Inductive Venn Predictors Based on Distance Learning

* IEEE International Conference on Omni-Layer Intelligent Systems (COINS), 2021, pp. 1-7
* Published in IEEE International Conference on Omni-Layer Intelligent Systems (COINS), 2021

Access Paper or Ask Questions

Improving Prediction Confidence in Learning-Enabled Autonomous Systems

Oct 07, 2021
Dimitrios Boursinos, Xenofon Koutsoukos

Figure 2 for Improving Prediction Confidence in Learning-Enabled Autonomous Systems

Figure 3 for Improving Prediction Confidence in Learning-Enabled Autonomous Systems

Figure 4 for Improving Prediction Confidence in Learning-Enabled Autonomous Systems

* Dynamic Data Driven Applications Systems. DDDAS 2020. Lecture Notes in Computer Science, vol 12312. Springer
* Published in Dynamic Data Driven Applications Systems. DDDAS 2020

Access Paper or Ask Questions

Assurance Monitoring of Learning Enabled Cyber-Physical Systems Using Inductive Conformal Prediction based on Distance Learning

Oct 07, 2021
Dimitrios Boursinos, Xenofon Koutsoukos

* Artificial Intelligence for Engineering Design, Analysis and Manufacturing, 35(2), 2021, 251-264
* Published in Artificial Intelligence for Engineering Design, Analysis and Manufacturing. arXiv admin note: text overlap with arXiv:2001.05014

Access Paper or Ask Questions

Detection of Dataset Shifts in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

Apr 14, 2021
Feiyang Cai, Ali I. Ozdagli, Xenofon Koutsoukos

Figure 2 for Detection of Dataset Shifts in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

Figure 3 for Detection of Dataset Shifts in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

Figure 4 for Detection of Dataset Shifts in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

* Accepted by the 4th IEEE International Conference on Industrial Cyber-Physical Systems (ICPS2021)

Access Paper or Ask Questions

Byzantine Resilient Distributed Multi-Task Learning

Oct 25, 2020
Jiani Li, Waseem Abbas, Xenofon Koutsoukos

Figure 2 for Byzantine Resilient Distributed Multi-Task Learning

Figure 3 for Byzantine Resilient Distributed Multi-Task Learning

Figure 4 for Byzantine Resilient Distributed Multi-Task Learning

* Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS 2020)

Access Paper or Ask Questions

Detecting Adversarial Examples in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

Mar 21, 2020
Feiyang Cai, Jiani Li, Xenofon Koutsoukos

Figure 2 for Detecting Adversarial Examples in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

Figure 3 for Detecting Adversarial Examples in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

Figure 4 for Detecting Adversarial Examples in Learning-Enabled Cyber-Physical Systems using Variational Autoencoder for Regression

* Accepted by Workshop on Assured Autonomous Systems (WAAS2020). arXiv admin note: text overlap with arXiv:2001.10494

Access Paper or Ask Questions