“Gradient Matching Generative Networks for Zero-Shot Learning”
Mert Bulent Sariyildiz (Bilkent University), Ramazan Gokberk
Cinbis (METU)
“Cross-task weakly supervised learning from instructional
videos” Dmitry Zhukov (Inria), Jean-Baptiste Alayrac (Google
DeepMind), Ramazan Gokberk Cinbis (METU), David Fouhey
(University of Michigan), Ivan Laptev (Inria), Josef Sivic
(Inria)
New MSc Thesis -
Edanur Demir successfully defended her MSc thesis titled
“Early-exit convolutional neural networks.” She was advised by
Emre Akbas.
New MSc Thesis -
Baran Baris Kivilcim successfully defended his MSc thesis titled
“Modeling Brain Networks with Artificial Neural Networks.” He
was advised by Fatos Yarman Vural.
New paper accepted to the Continual Learning Workshop at NeurIPS
2018. -
“Learning
to Increment a Contextual Model” by Fethiye Irmak Doğan,
Hande Çelikkanat, İlker Bozcan, Sinan Kalkan.
New paper at the Robotics and Autonomous Systems Journal. -
“COSMO:
Contextualized scene modeling with Boltzmann Machines” by
İlker Bozcan, Sinan Kalkan.
İlker Bozcan’s thesis received the Best Thesis award. -
“Best Thesis
of the Year”. İlker’s thesis was advised by Sinan Kalkan.
New book translation -
Goodfellow et al.’s book “Deep Learning” has been translated
into Turkish by H. Aydın, R. G. Cinbiş, Y. D. Çetin, B. Demirel,
S. Kalkan, H. Moğultay, M. B. Sarıyıldız, G. Sümbül, F. Yarman
Vural Buzdağı Yayınevi, 2018
New paper accepted to the Visual Computer Journal. -
“Deep
3D semantic scene extrapolation” Ali Abbasi, Sinan Kalkan,
Yusuf Sahillioğlu.
Two ECCV2018, one BMVC2018 and one IROS2018 papers accepted! -
New MSc Thesis -
Muhammed Kocabas successfully defended his MSc thesis titled
“2D/3D human pose estimation using deep convolutional neural
networks.” He was advised by Emre Akbas.
