| A comparison of loss weighting strategies for multi task learning in deep neural networks T Gong, T Lee, C Stephenson, V Renduchintala, S Padhy, A Ndirango, ... IEEE Access 7, 141627-141632, 2019 | 144 | 2019 |
| Wireline receiver circuitry having collaborative timing recovery T Musah, G Keskin, G Balamurugan, JE Jaussi, BK Casper US Patent 9,374,250, 2016 | 113 | 2016 |
| A 4–32 Gb/s bidirectional link with 3-tap FFE/6-tap DFE and collaborative CDR in 22 nm CMOS T Musah, JE Jaussi, G Balamurugan, S Hyvonen, TC Hsueh, G Keskin, ... IEEE Journal of Solid-State Circuits 49 (12), 3079-3090, 2014 | 67 | 2014 |
| Compression in machine learning and deep learning processing J Ray, B Ashbaugh, P Surti, P Ramani, R Harihara, JC Justin, J Huang, ... US Patent 10,546,393, 2020 | 60 | 2020 |
| Mismatch analysis and statistical design at 65 nm and below L Pileggi, G Keskin, X Li, K Mai, J Proesel 2008 IEEE Custom Integrated Circuits Conference, 9-12, 2008 | 50 | 2008 |
| An 8-bit 1.5 GS/s flash ADC using post-manufacturing statistical selection J Proesel, G Keskin, JO Plouchart, L Pileggi IEEE Custom Integrated Circuits Conference 2010, 1-4, 2010 | 44 | 2010 |
| 26.4 A 25.6 Gb/s differential and DDR4/GDDR5 dual-mode transmitter with digital clock calibration in 22nm CMOS TC Hsueh, G Balamurugan, J Jaussi, S Hyvonen, J Kennedy, G Keskin, ... 2014 IEEE International Solid-State Circuits Conference Digest of Technical …, 2014 | 40 | 2014 |
| Improving branch prediction by modeling global history with convolutional neural networks SJ Tarsa, CK Lin, G Keskin, G Chinya, H Wang arXiv preprint arXiv:1906.09889, 2019 | 35 | 2019 |
| Statistical modeling and post manufacturing configuration for scaled analog CMOS G Keskin, J Proesel, L Pileggi IEEE Custom Integrated Circuits Conference 2010, 1-4, 2010 | 35 | 2010 |
| Exploiting combinatorial redundancy for offset calibration in flash ADCs G Keskin, J Proesel, JO Plouchart, L Pileggi IEEE journal of solid-state circuits 46 (8), 1904-1918, 2011 | 33 | 2011 |
| Redat: Accent-invariant representation for end-to-end asr by domain adversarial training with relabeling H Hu, X Yang, Z Raeesy, J Guo, G Keskin, H Arsikere, A Rastrow, ... ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and …, 2021 | 31 | 2021 |
| 26.2 A 205mW 32Gb/s 3-Tap FFE/6-tap DFE bidirectional serial link in 22nm CMOS J Jaussi, G Balamurugan, S Hyvonen, TC Hsueh, T Musah, G Keskin, ... 2014 IEEE International Solid-State Circuits Conference Digest of Technical …, 2014 | 31 | 2014 |
| Runtime processor optimization SJ Tarsa, GN Chinya, G Keskin, H Wang, K Sankaranarayanan US Patent App. 15/444,390, 2018 | 13 | 2018 |
| Measuring the Effectiveness of Voice Conversion on Speaker Identification and Automatic Speech Recognition Systems G Keskin, T Lee, C Stephenson, OH Elibol International Conference on Machine Learning 2019, Synthetic Realities …, 2019 | 12 | 2019 |
| Adversarially trained autoencoders for parallel-data-free voice conversion O Ocal, OH Elibol, G Keskin, C Stephenson, A Thomas, K Ramchandran ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and …, 2019 | 12 | 2019 |
| Supporting learned branch predictors G Keskin, SJ Tarsa, GN Chinya, TH Lin, PH Wang, H Wang US Patent 10,534,613, 2020 | 11 | 2020 |
| Wireline receiver circuitry having collaborative timing recovery T Musah, G Keskin, G Balamurugan, JE Jaussi, BK Casper US Patent 9,794,089, 2017 | 10 | 2017 |
| Active on-die suppression of power supply noise G Keskin, X Li, L Pileggi IEEE Custom Integrated Circuits Conference 2006, 813-816, 2006 | 10 | 2006 |
| Self-Healing Circuits Using Statistical Element Selection. G Keskin Carnegie Mellon University, 2010 | 7 | 2010 |
| Prune then distill: Dataset distillation with importance sampling AS Sundar, G Keskin, C Chandak, IF Chen, P Ghahremani, S Ghosh ICASSP 2023-2023 IEEE International Conference on Acoustics, Speech and …, 2023 | 6 | 2023 |
