Berkay Kicanaoglu

Research Scientist at Flawless AI ⋅ Neural Rendering ⋅ PhD in ML/CV @University of Amsterdam

[Berkay Kicanaoglu @KicanaogluB]

@umutsimsekli @ERC_Research @inria_paris @Sierra_ML_Lab Tebrikler 💫✨

[Tolga Birdal @tolga_birdal]

Signed, sealed, delivered. I am beyond excited to be joining @imperialcollege @ICComputing as an assistant professor (lecturer) in 2022. As a group, we will focus on 3D computer vision, geometric machine learning and topological data analysis. #CVPR #CVML #AI #ML #MachineLearning

[Berkay Kicanaoglu @KicanaogluB]

@tolga_birdal @imperialcollege @ICComputing Amazing 🤩 Congrats and best wishes ✌🏼

[Berkay Kicanaoglu @KicanaogluB]

@hen_drik @andy2pandy @roboticwrestler @Harvard @Rolf_Drechsler @cbpuschmann @gabymlp Congratulations, Hendrik! 🥳

[Berkay Kicanaoglu @KicanaogluB]

Generating images via sequential updates published at AAAI’19. @0ozgur0’s thesis work last year. w/ @twuilliam

[Ozgur Turkoglu @0ozgur0]

Check out my post on sequential images generation with GANs based on our AAAI19 paper with @twuilliam @KicanaogluB https://t.co/xAhjl60z1i

[Sanja Fidler @FidlerSanja]

Current image datasets are getting very large and pre-training on them is time consuming. We are releasing Neural Data Server (NDS), a search engine for transfer learning data! @yanxi0830 @davidjesusacu #UofT
Webservice: https://t.co/3vg5uRlk2U
Paper: https://t.co/OEFWJKHETH

[Maurice Weiler @maurice_weiler]

Our work on Gauge Equivariant CNNs made it into Quanta Magazine!
The article gives a nice overview on coordinate independent convolutions and connections between theoretical physics and deep learning.

[Maurice Weiler @maurice_weiler]

Check out our poster #143 on general E(2)-Steerable CNNs tomorrow, Thu 10:45AM.
Our work solves for the most general isometry-equivariant convolutional mappings and implements a wide range of related work in a unified framework.
With @_gabrielecesa_
#NeurIPS2019 #NeurIPS

[Kosta Derpanis (sabbatical in Zurich) @CSProfKGD]

Kaolin: A #PyTorch Library for Accelerating 3D #DeepLearning Research

Paper: https://t.co/ObmdZAxfcQ

Project page: https://t.co/IInSnHKeGM

[Ivan Sosnovik @isosnovik]

Check out "Scale-Equivariant Steerable Networks"
(https://t.co/deebcUIh6R).
It is joint work with Michał Szmaja and Arnold Smeulders.
We build scale-equivariant CNNs which do not use image rescaling and do not limit the admissible scale factors.

[Carl Allen @carl_s_allen]

Delighted to receive #bestpaper Honourable Mention @icmlconf and now finish my blog post! Explaining why king - man + woman = queen: https://t.co/TVgL8fiDC7 … #icml2019 #word2vec #glove @tmh31 @yoavgo @omerlevy_ @JeffDean @RichardSocher

[MIT Technology Review @techreview]

This is how artificial intelligence can help save our planet and humanity from climate change. https://t.co/xYwCavSf3b

[Tolga Birdal @tolga_birdal]

Happy to hear that our work with @umutsimsekli qualifies as a best paper finalist @cvpr2019.
https://t.co/P6m6kV3rm1

[Taylan Cemgil @TaylanCemgilML]

Three great papers and excellent talks from @umutsimsekli and https://t.co/H2FVNks09l. in #ICML2019. In just an hour. The last photo shows the question queue for the award winning talk with @leventsagun and Mert Gurbuzbalaban :-)

[Taylan Cemgil @TaylanCemgilML]

Umut Simsekli, @umutsimsekli Levent Sagun @leventsagun ve Mert Gürbüzbalaban. #icml2019 honorable mention! Congratulations!

[Maurice Weiler @maurice_weiler]

Come to our talk on Gauge Equivariant Convolutional Networks and Icosahedral CNNs today at 14:40 @ Grand Ballroom, #ICML2019.
Happy to discuss more details and connections to physics at poster #76 @ Pacific Ballroom, 18:30.
With @TacoCohen, @KicanaogluB and @wellingmax .

[Daniel Worrall @danielewworrall]

**NEW WORK** My masters student @diacon995 and I learn how to (group) convolve in our new paper "Learning to Convolve": https://t.co/obCv5xQ9wT accepted to @icmlconf. Simple idea with good results. TAKE HOME: Learned group equivariant weight basis > standard pixel basis

[Berkay Kicanaoglu @KicanaogluB]

Hyviä uutisia Suomesta!

[Cees Snoek @cgmsnoek]

Finnish #AI course passes 140K students. #bravo https://t.co/GwhoknqVtz @JimStolze @Ai_Cursus

[Taco Cohen @TacoCohen]

Interested in geometric and equivariant deep learning? Check out our latest paper on Gauge Equivariant CNNs, where we show how gauge theory makes it possible to build CNNs on general manifolds: https://t.co/NAZGve9YnP

[Berkay Kicanaoglu @KicanaogluB]

@bebusag @cumhurk7 📿📿