Berkin Akin

Berkin Akin

I am a Software Engineer at Google DeepMind where I currently focus on compute performance for large-scale AI models deployed on TPUs, previously I focused on small-scale AI models deployed on Edge TPUs. Prior to joining Google in 2019, I was a research scientist at Intel Labs. Before that I got my PhD from Carnegie Mellon University.

Research Areas

Authored Publications
0 results

Sort By
  • Title
  • Title, descending
  • Year
  • Year, descending
    Google
Searching for Efficient Neural Architectures for On-Device ML on Edge TPUs
Anton Spiridonov
Berkin Akin
Hao Xu
Marie Charisse White
Ping Zhou
Suyog Gupta
Yanqi Zhou
Yun Long
Zhuo Wang
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) (2022)
An Evaluation of Edge TPU Accelerators for Convolutional Neural Networks
Amir Yazdanbakhsh
Berkin Akin
Kiran K Seshadri
ArXiv, https://arxiv.org/abs/2102.10423 (2021)
Accelerator-aware Neural Network Design using AutoML
Berkin Akin
Suyog Gupta
On-device Intelligence Workshop at MLSys 2020 (2020)
Apollo: Transferable Architecture Exploration
Albin Jones
Amir Yazdanbakhsh
Berkin Akin
Christof Angermueller
James Pierce Laudon
Kevin Swersky
Milad Hashemi
Ravi Narayanaswami
Sat Chatterjee
Yanqi Zhou
ML for Systems Workshop at NeurIPS 2020
Search on Google Scholar