Picture for Ajinkya Tejankar

Ajinkya Tejankar

Defending Against Patch-based Backdoor Attacks on Self-Supervised Learning

Apr 04, 2023
Ajinkya Tejankar, Maziar Sanjabi, Qifan Wang, Sinong Wang, Hamed Firooz, Hamed Pirsiavash, Liang Tan

Figure 1 for Defending Against Patch-based Backdoor Attacks on Self-Supervised Learning

Figure 2 for Defending Against Patch-based Backdoor Attacks on Self-Supervised Learning

Figure 3 for Defending Against Patch-based Backdoor Attacks on Self-Supervised Learning

Figure 4 for Defending Against Patch-based Backdoor Attacks on Self-Supervised Learning

* Accepted to CVPR 2023

Access Paper or Ask Questions

SimReg: Regression as a Simple Yet Effective Tool for Self-supervised Knowledge Distillation

Jan 13, 2022
K L Navaneet, Soroush Abbasi Koohpayegani, Ajinkya Tejankar, Hamed Pirsiavash

Figure 1 for SimReg: Regression as a Simple Yet Effective Tool for Self-supervised Knowledge Distillation

Figure 2 for SimReg: Regression as a Simple Yet Effective Tool for Self-supervised Knowledge Distillation

Figure 3 for SimReg: Regression as a Simple Yet Effective Tool for Self-supervised Knowledge Distillation

Figure 4 for SimReg: Regression as a Simple Yet Effective Tool for Self-supervised Knowledge Distillation

* In BMVC 2021. Code available at: https://github.com/UCDvision/simreg

Access Paper or Ask Questions

Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning

Dec 08, 2021
Ajinkya Tejankar, Soroush Abbasi Koohpayegani, KL Navaneet, Kossar Pourahmadi, Akshayvarun Subramanya, Hamed Pirsiavash

Figure 1 for Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning

Figure 2 for Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning

Figure 3 for Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning

Figure 4 for Constrained Mean Shift Using Distant Yet Related Neighbors for Representation Learning

* Code is available at https://github.com/UCDvision/CMSF. arXiv admin note: text overlap with arXiv:2110.10309

Access Paper or Ask Questions