About Me

I'm the Co-Founder and CEO of Deci, a Deep learning AutoML focused startup. Prior to that, I've done my PhD at the Computer Science Department at the Technion, working under the supervision of Prof. Ran El-Yaniv. My research centers around the uncertainty of machine learning algorithms and exposes me to the limitations of machines as agents of knowledge and intelligence. My aim is to enhance the reliability of prediction by developing methods for estimation and control of uncertainty in machine learning algorithms.

Recent Work

Ran El-Yaniv, Yonatan Geifman, Yair Wiener. The Prediction Advantage: A Universally Meaningful Performance Measure for Classification and Regression - preprint arXiv, Slides

Yonatan Geifman and Ran El-Yaniv. Selective Classification for Deep Neural Networks. In Advances in Neural Information Processing Systems, 2017. arXiv, code, Slides

Yonatan Geifman and Ran El-Yaniv. Deep Active Learning over the Long Tail - preprint arXiv

Yonatan Geifman, Guy Uziel, Ran El-Yaniv. Bias-Reduced Uncertainty Estimation for Deep Neural Classifiers - ICLR 2019 arXiv, code

Yonatan Geifman and Ran El-Yaniv. SelectiveNet: A Deep Neural Network with an Integrated Reject Option - ICML 2019 arXiv code

Yonatan Geifman and Ran El-Yaniv. Deep Active Learning with a Neural Architecture Search - NeurIPS 2019 arXiv

Talks

Deep Active Learning with a Neural Architecture Search - NeurIPSi meetup Tel-Aviv (November 2019) - slides

Uncertainty Estimation and its applications in deep neural networks - IBM research (August 2019) - slides

SelectiveNet: A Deep Neural Network with an Integrated Reject Option - ICML 2019 - ICML oral presentation (June 2019) - slides

Uncertainty Estimation and its applications in deep neural networks - Phd seminar talk (June 2019) - slides

Uncertainty and its applications in deep neural networks - Trigo Vision meetup (2019) - slides

On the estimation and control of uncertainty in deep neural networks - Amazon (2018) - slides

Teaching

Winter 2018-2019 - Deep Learning, 236606, TA in charge

Spring 2018 - Deep Learning, 236606, TA in charge

Winter 2017-2018 - [Introduction to computer science with Matlab], TA in charge

Spring 2017 - [Introduction to computer science], TA

Winter 2016-2017 - [Introduction to computer science with Matlab], TA

Experience

Sep 2019 - Co-Founder and CEO at Deci a Deep learning AutoML focused startup.

Summer 2018 - Intern at Google AI (Mountain view) - Machine perception team - Morphnet

Contact

Feel free to contact me at yonatan.g at cs dot technion dot ac dot il