Symbolic Density Models of One-in-a-Billion Statistical Tails via Importance Sampling and Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.3949

@InCollection{McConaghy:2010:GPTP,
  author =       "Trent McConaghy",
  title =        "Symbolic Density Models of One-in-a-Billion
                 Statistical Tails via Importance Sampling and Genetic
                 Programming",
  booktitle =    "Genetic Programming Theory and Practice VIII",
  year =         "2010",
  editor =       "Rick Riolo and Trent McConaghy and 
                 Ekaterina Vladislavleva",
  series =       "Genetic and Evolutionary Computation",
  volume =       "8",
  address =      "Ann Arbor, USA",
  month =        "20-22 " # may,
  publisher =    "Springer",
  chapter =      "10",
  pages =        "161--173",
  keywords =     "genetic algorithms, genetic programming, symbolic
                 regression, density estimation, importance sampling,
                 Monte Carlo methods, memory, SRAM, integrated circuits,
                 extreme-value statistics",
  isbn13 =       "978-1-4419-7746-5",
  URL =          "http://www.springer.com/computer/ai/book/978-1-4419-7746-5",
  URL =          "http://trent.st/content/2010-GPTP-tails.pdf",
  size =         "14 pages",
  abstract =     "This paper explores the application of symbolic
                 regression for building models of probability
                 distributions in which the accuracy at the
                 distributions' tails is critical. The problem is of
                 importance to cutting-edge industrial integrated
                 circuit design, such as designing SRAM memory
                 components (bitcells, sense amps) where each component
                 has extremely low probability of failure. A naive
                 approach is infeasible because it would require
                 billions of Monte Carlo circuit simulations. This paper
                 demonstrates a flow that efficiently generates samples
                 at the tails using importance sampling, then builds
                 genetic programming symbolic regression models in a
                 space that captures the tails, the normal quantile
                 space. These symbolic density models allow the circuit
                 designers to analyse the tradeoff between high-sigma
                 yields and circuit performance. The flow is validated
                 on two modern industrial problems: a bitcell circuit on
                 a 45nm TSMC process, and a sense amp circuit on a 28nm
                 TSMC process.",
  notes =        "part of \cite{Riolo:2010:GPTP}",
}

Genetic Programming entries for Trent McConaghy

Citations