The ACIP Project


Cognitive computing systems make intelligent decisions, learn from previous experiences, and are introspective, constantly improving their behavior and performance over time. Recently, the Artificial Intelligence community has made significant advances in algorithms for cognitive processing. Areas of innovation include Bayesian Inference, Evolutionary Computing, and Knowledge-Based Reasoning. The ACIP project brings together algorithm designers and computer architects to develop high-performance implementations of cognitive systems. Key areas of innovation span algorithms, compilers, runtime systems, and architectures.

The ACIP project is a DARPA initiative, and involves researchers from academia, industry, and the Department of Defense. At UMD, we are conducting an in-depth applications study to identify the performance requirements of key cognitive kernels. Our group is also developing architectural support to speed up these kernels. Current techniques under investigation include mechanisms for exploiting soft computing properties, data-driven processor architectures, computation and data migration for locality management, and helper threading for introspection and dynamic optimization.



  • Donald Yeung
  • Students

  • Hameed Badawy
  • Xuanhua Li
  • Wanli Liu
  • Priyanka Rajkhowa
  • Meng-Ju Wu
  • Publications:

  • Xuanhua Li and Donald Yeung. Application-Level Correctness and its Impact on Fault Tolerance. To appear in Proceedings of The 13th International Symposium on High-Performance Computer Architecture (HPCA-XIII). Phoenix, AZ. February 2007. (pdf, postscript)
  • Xuanhua Li and Donald Yeung. Exploiting Soft Computing for Increased Fault Tolerance. Appears in Proceedings of the 2006 Workshop on Architectural Support for Gigascale Integration(ASGI'06). co-held with the ISCA-33. Boston, MA. June 2006. (pdf, postscript)
  • Funding:

    The ACIP project is funded by the Defense Advanced Research Projects Agency (DARPA) through the Department of the Interior National Business Center under grant #NBCH104009.

    Last updated: September 2006 by Donald Yeung (