Valeriu Codreanu
Home * People * Valeriu Codreanu
Valeriu Codreanu,
a Romanian computer scientist, electrical engineer and head of high-performance computing and visualization at SURFsara, Amsterdam Science Park, part of the SURF cooperative association of Dutch educational and research institutions in the field of information and communication technology.
He holds B.Sc., M.Sc. and Ph.D. degrees from Politehnica University of Bucharest in 2003, 2008 and 2011 respectively,
and was postdoctoral researcher at Eindhoven University of Technology and University of Groningen, working on GPU computing, computer vision, and embedded systems [2].
His research interest include computer architecture, in particular multi/many core architectures, deep learning, and compilation techniques for state-of-the-art parallel processors, including auto parallelization and auto vectorization [1].
As co-advisor of Matthia Sabatelli concerning his M.Sc. thesis on learning to play chess with minimal lookahead,
using multilayer perceptrons versus convolutional neural networks [3], Valeriu Codreanu co-authored the subsequent ICPRAM paper Learning to Evaluate Chess Positions with Deep Neural Networks and Limited Lookahead [4].
Selected Publications
2013 ...
- David Williams, Valeriu Codreanu, Po Yang, Baoquan Liu, Feng Dong, Burhan Yasar, Babak Mahdian, Alessandro Chiarini, Xia Zhao, Jos Roerdink (2013). Evaluation of Autoparallelization Toolkits for Commodity GPUs. PPAM 2013
- Baoquan Liu, Alexandru Telea, Jos Roerdink, Gordon Clapworthy, David Williams, Po Yang, Feng Dong, Valeriu Codreanu, Alessandro Chiarini (2018). Parallel centerline extraction on the GPU. Computers & Graphics, Vol. 41, pdf
- Cedric Nugteren, Valeriu Codreanu (2017). CLTune: A Generic Auto-Tuner for OpenCL Kernels. arXiv:1703.06503
- Valeriu Codreanu, Damian Podareanu, Vikram Saletore (2017). Scale out for large minibatch SGD: Residual network training on ImageNet-1K with improved accuracy and reduced time to train. arXiv:1711.04291
- Matthia Sabatelli, Francesco Bidoia, Valeriu Codreanu, Marco Wiering (2018). Learning to Evaluate Chess Positions with Deep Neural Networks and Limited Lookahead. ICPRAM 2018, pdf
- Ananth Sankaranarayanan, Valeriu Codreanu, Damian Podareanu, Colin Healy (2018). Efficient neural network training on Intel Xeon-based supercomputers. AI Conference 2018
2020 ...
- Caspar van Leeuwen, Damian Podareanu, Valeriu Codreanu, Maxwell X. Cai, Axel Berg, Simon Portegies Zwart, Robin Stoffer, Menno Veerman, Chiel van Heerwaarden, Sydney Otten, Sascha Caron, Cunliang Geng, Francesco Ambrosetti, Alexandre M.J.J. Bonvin (2020). Deep-learning enhancement of large scale numerical simulations. arXiv:2004.03454
- Florian Rehm, Sofia Vallecorsa, Vikram Saletore, Hans Pabst, Adel Chaibi, Valeriu Codreanu, Kerstin Borras, Dirk Krücker (2021). Reduced Precision Strategies for Deep Learning: A High Energy Physics Generative Adversarial Network Use Case. arXiv:2103.10142
External Links
- Valeriu Codreanu | LinkedIn
- Intel® Parallel Computing Center at SURFsara BV, 2018
- Interview: SURF pushes the boundaries of deep learning | SURF.nl, 2018
References
- ↑ 1.0 1.1 Valeriu Codreanu | LinkedIn
- ↑ Speaker: Valeriu Codreanu: Artificial Intelligence Conference: Applied AI & machine learning
- ↑ Matthia Sabatelli (2017). Learning to Play Chess with Minimal Lookahead and Deep Value Neural Networks. Master's thesis, University of Groningen, pdf
- ↑ Matthia Sabatelli, Francesco Bidoia, Valeriu Codreanu, Marco Wiering (2018). Learning to Evaluate Chess Positions with Deep Neural Networks and Limited Lookahead. ICPRAM 2018, pdf
- ↑ dblp: Valeriu Codreanu