Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Dimche Kostadinov

Saliency Diversified Deep Ensemble for Robustness to Adversaries


Dec 07, 2021
Alex Bogun, Dimche Kostadinov, Damian Borth

* Accepted to AAAI Workshop on Adversarial Machine Learning and Beyond 2022 

  Access Paper or Ask Questions

Self-Supervised Representation Learning on Neural Network Weights for Model Characteristic Prediction


Nov 03, 2021
Konstantin Schürholt, Dimche Kostadinov, Damian Borth

* Published at 35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia. 31 Pages, 14 figures 

  Access Paper or Ask Questions

Zero-shot Voice Conversion via Self-supervised Prosody Representation Learning


Oct 27, 2021
Shijun Wang, Dimche Kostadinov, Damian Borth


  Access Paper or Ask Questions

Privacy-Preserving Near Neighbor Search via Sparse Coding with Ambiguation


Feb 08, 2021
Behrooz Razeghi, Sohrab Ferdowsi, Dimche Kostadinov, Flavio. P. Calmon, Slava Voloshynovskiy

* To be presented at 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2021) 

  Access Paper or Ask Questions

Unsupervised Feature Learning for Event Data: Direct vs Inverse Problem Formulation


Sep 30, 2020
Dimche Kostadinov, Davide Scaramuzza

* IAPR IEEE/Computer Society International Conference on Pattern Recognition (ICPR), Milan, 2021 

  Access Paper or Ask Questions

Online Weight-adaptive Nonlinear Model Predictive Control


Aug 06, 2020
Dimche Kostadinov, Davide Scaramuzza

* IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Las Vegas, 2020 

  Access Paper or Ask Questions

Network Parameter Learning Using Nonlinear Transforms, Local Representation Goals and Local Propagation Constraints


Jan 31, 2019
Dimche Kostadinov, Behrooz Razdehi, Slava Voloshynovskiy

* arXiv admin note: text overlap with arXiv:1805.07802 

  Access Paper or Ask Questions

Clustering with Jointly Learned Nonlinear Transforms Over Discriminating Min-Max Similarity/Dissimilarity Assignment


Jan 30, 2019
Dimche Kostadinov, Behrooz Razeghi, Taras Holotyak, Slava Voloshynovskiy


  Access Paper or Ask Questions

Network Learning with Local Propagation


May 20, 2018
Dimche Kostadinov, Behrooz Razeghi, Sohrab Ferdowsi, Slava Voloshynovskiy

* preprint, a similar version submitted to NIPS 2018 

  Access Paper or Ask Questions

A multi-layer network based on Sparse Ternary Codes for universal vector compression


Oct 31, 2017
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov

* Submitted to ICASSP 2018 

  Access Paper or Ask Questions

Privacy Preserving Identification Using Sparse Approximation with Ambiguization


Sep 29, 2017
Behrooz Razeghi, Slava Voloshynovskiy, Dimche Kostadinov, Olga Taran

* submitted to WIFS 2017 

  Access Paper or Ask Questions

A multi-layer image representation using Regularized Residual Quantization: application to compression and denoising


Jul 07, 2017
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov

* At the International Conference on Image Processing 2017 (ICIP'17), Beijing, China 

  Access Paper or Ask Questions

Regularized Residual Quantization: a multi-layer sparse dictionary learning approach


May 01, 2017
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov

* To be presented at SPARS 2017, Lisbon, Portugal 

  Access Paper or Ask Questions

Sparse Ternary Codes for similarity search have higher coding gain than dense binary codes


Apr 25, 2017
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov, Taras Holotyak

* Accepted at 2017 IEEE International Symposium on Information Theory (ISIT'17) 

  Access Paper or Ask Questions

Sparse Multi-layer Image Approximation: Facial Image Compression


Jun 12, 2015
Sohrab Ferdowsi, Svyatoslav Voloshynovskiy, Dimche Kostadinov

* Submitted to the MLSP 2015 

  Access Paper or Ask Questions