Bilge Acun


CV: [pdf]
Google Scholar
Github Twitter

About me

Bilge Acun is a Research Scientist at Meta AI / FAIR, SysML team. She is working on making large scale machine learning systems more efficient through algorithmic and system optimizations.


She received her Ph.D. degree in 2017 at the Department of Computer Science at University of Illinois at Urbana-Champaign, advised by Professor Laxmikant V. Kale. Before joining Meta/Facebook, she worked at the IBM Thomas J. Watson Research Center as a Research Staff Member.


Her research interests include Systems for Machine Learning, Parallel and Distributed Computing, Computer Architecture, Energy Efficient Computing, Smart Runtime Systems.

Selected Publications [Full List]

Sustainable Datacenters and AI

[ASPLOS'23] [IEEE Micro Top Picks'24 Honorable Mention] [paper]  [bib] [code] [talk at Stanford] 

Carbon Explorer: A Holistic Approach for Designing Carbon Aware Datacenters

B. Acun, B. Lee, F. Kazhamiaka, K. Maeng, M. Chakkaravarthy, U. Gupta, D. Brooks, C. J. Wu
ACM International Conference on Architectural Support for Programming Languages and Operating Systems, 2023.

[MLSys'22]  [paper]  [bib]

Sustainable AI: Environmental Implications, Challenges and Opportunities

C. J. Wu, R. Raghavendra, U. Gupta, B. Acun, et al.
Conference on Machine Learning and Systems, 2022.

Efficient ML Training and Inference

[ICML'24]  [paper]  [bib]

CHAI: Clustered Head Attention for Efficient LLM Inference

S. Agarwal, B. Acun, B. Homer, M. Elhoushi, Y. Lee, S. Venkataraman, D. Papailiopoulos, C.-J. Wu
International Conference on Machine Learning, 2024.

[MLSys'21] [Outstanding Paper Award] [paper]  [bib] [code] 

TT-Rec: Tensor Train Compression for Deep Learning Recommendation Models

C. Yin, B. Acun, X. Liu, C. J. Wu
Conference on Machine Learning and Systems,2021.

[HPCA'21]  [paper] [bib]

Understanding Training Efficiency of Deep Learning Recommendation Models at Scale

B. Acun, M. Murphy, X. Wang, J. Nie, C. J. Wu, K. Hazelwood
IEEE International Symposium on High-Performance Computer Architecture, 2021.

[Micro'21]  [paper] [bib]

Datacenter-Scale Analysis and Optimization of GPU Machine Learning Workloads

L. Wesolowski, B. Acun, V. Andrei, A. Aziz, G. Dankel, C. Gregg, X. Meng, C. Meurillon, D. Sheahan, L. Tian, J. Yang, P. Yu, K. Hazelwood
IEEE Micro, 2021.

Systems & Computer Architecture

[HPCA'19]  [paper] [bib]

Power-Aware Heterogeneous Node Assembly

B. Acun, A Buyuktosunoglu, E. K. Lee, Y. Park
IEEE International Symposium on High-Performance Computer Architecture, 2019.

[ICS'16]  [paper] [bib]

Variation Among Processors Under Turbo Boost in HPC Systems

B. Acun , P. Miller, L. V. Kalé
International Conference on Supercomputing, 2016.

[Computer'16]  [Cover Featured] [paper] [bib]

Power, Reliability, Performance: One System to Rule Them All

B. Acun, A. Langer, H. Menon, O. Sarood, E. Totoni, and L. V. Kalé.
IEEE Computer, Energy Efficient Computing Special Issue, 2016.

[SC'14]  [paper] [bib] [code]

Parallel Programming with Migratable Objects: Charm++ in Practice

B. Acun, A. Gupta, N. Jain, A. Langer, H. Menon, E. Mikida, X. Ni, M. Robson, Y. Sun, E. Totoni, L. Wesolowski, L. V. Kalé.
Supercomputing (SC), 2014.

Honors and Awards

Patents


  1. Device Identification via Chip Manufacturing Related Fingerprints

    B. Acun
    US Patent No: 11,205,018, 2019.
  2. Locality Aware Data Loading for Machine Learning

    C. C. Yang, G. Cong, B. Acun, A. Morari
    US Patent No: 11,093,862, 2019.
  3. Learning-based Thermal Estimation in Multicore Architectures

    E. K. Lee, B. Acun, Y. Park, P. W. Coteus
    US Patent No: 11,334,398, 2018.
  4. Variation-Aware Intra-Node Power Shifting Among Different Hardware Components

    E. K. Lee, B. Acun, Y. Park, A. Morari, A. Buyuktosunoglu
    US Patent No: 10,761,583, 2018.
  5. Job Scheduling Based on Node and Application Characteristics

    B. Acun, E. K. Lee, Y. Park
    US Patent No: 10,725,834, 2017.
  6. Power Efficiency Aware Node Component Assembly

    B. Acun, E. K. Lee, Y. Park
    US Patent No: 10,831,252, 2017.
  7. Systems and Methods for Computer Input

    B. Acun, E. Candan
    US Patent App. No: 15/252,163, 2017.

Services and Activities

Organizing Activities


  • PC Chair for the HPC for Machine Learning track at Supercomputing (SC'24)

  • PC Member at Conference on Machine Learning and Systems (MLSys'22 & '23 & '24)

  • PC Member at IEEE International Parallel & Distributed Processing Symposium (IPDPS'22 & IPDPS'23)

  • PC Member at Workshop on Benchmarking Machine Learning Workloads on Emerging Hardware (MLSys'21)

  • PC Member at the 37th IEEE International Conference on Computer Design (ICCD'19).

  • PC Member at the Women in High Performance Computing Workshop (WHPC at SC'18).

  • Workshop Organizer for “Ethics in Computing Workshop” at Heidelberg Forum (HLF'18).

  • Co-organizer, Women Empowered in STEM Conference (weSTEM'14).

Peer Review Activities

  • External Reviewer for IEEE Transactions on Parallel and Distributed Systems (TPDS'22).

  • External Reviewer for IEEE Transactions on Cloud Computing (TCC'19).

  • External Reviewer for Latin America High Performance Computing Conference (CARLA) (CARLA'19).

Other Activities

  • Technical Session Chair, "Data Centers" & "Energy Efficiency and Measurements" Sessions at IGSC'19.

  • Technical Session Chair, “Interfaces Session”, at 15th Annual Charm++ Workshop and Applications, Charm++'17.

  • Technical Session Chair, “Applications Session”, at 14th Annual Charm++ Workshop and Applications, Charm++'16.

  • Charm++ Hands-on Tutorial, at Argonne Training Program on Extreme-Scale Computing, ATPESC'15.

  • Graduate Student Mentor, Department of Computer Science at UIUC. 2013-2015.