Stochastic Optimization with Learning for Complex Problems
Continue the development of the learning framework and evaluate incremental learning approaches (Bayesian, etc.)
Apply the techniques to one or two new problems: standard-cell library generation, optimal clustering for placement
Evaluate the effectiveness of adaptive optimization on other problems: BDD variable sifting
SPECIFIC MILESTONES AND DELIVERABLES:
Complete standard-cell placement evaluation (Mar-97)
Implement adaptive approach to BDD variable sifting and evaluate (Jul-97)
Formulate approach to optimal clustering problem and implement prototype (Sep-97)
Implement incremental learning algorithms in the framework and evaluate (Dec-97)