dr Kamil Deja.
Adiunkt w CVLab, zakładu Grafiki Komputerowej , Instytutu Informatyki Politechniki Warszawskiej
Zainteresowania naukowe:
Uczenie Maszynowe, Modele Generatywne, Uczenie Ciągłe
Wybrane publikacje:
- Learning Data Representations with Joint Diffusion Models (ECML 2023) Pre-print
- Diffusion-based accent modelling in speech synthesis (Interspeech 2023) Pre-print
- On Analyzing Generative and Denoising Capabilities of Diffusion-based Deep Generative Models (NeurIPS 2022) Pre-print
- Automatic Evaluation of Speaker Similarity (Interspeech 2022) Pre-print
- Multiband VAE: Latent Space Alignment for Knowledge Consolidation in Continual Learning (IJCAI-ECAI 2022) Pre-print
- BinPlay: A Binary Latent Autoencoder for Generative Replay Continual Learning (IJCNN 2021) Pre-print
- End-to-end Sinkhorn Autoencoder with Noise Generator (IEEE Acess) Open access
