------------------------------------------------------------------------------- *** Note Day and Time *** *** Note Day and Time *** ------------------------------------------------------------------------------- B O S T O N U N I V E R S I T Y Computer Science Department C O L L O Q U I U M Tuesday, February 7, 1995 Janis Barzdins 9:30am (Coffee served at 9:00am) ** AND ** Rusins Freivalds 11:00am (Coffee served at 10:30am) Seminar Room / MCS 135 ------------------------------------------------------------------------------- TOWARDS EFFICIENT INDUCTIVE SYNTHESIS FROM INPUT/OUTPUT EXAMPLES Janis Barzdins Institute of Mathematics and Computer Science, University of Latvia 9:30am We will begin with recursive - theoretic approach to Inductive Synthesis. Then more practical approaches will be considered. Inductive Synthesis method based on finding local regularities (algebraic axioms) will be discussed. Special attention will be paid to the synthesis of expressions from input/output examples. Efficient Inductive Synthesis algorithm will be presented. Corresponding computer experiments will be described. At the conclusion efficient Inductive Synthesis of Term Rewriting Systems will be considered, including computer experiments. Our conclusion will say that the synthesis of nontrivial functions and algorithms in a realistic time, entirely from input/output examples, is not a hopeless task. LOWER BOUNDS OF SPACE AND TIME COMPLEXITY OF RANDOMIZED MACHINES Rusins Freivalds University of Latvia 11:00am Randomized and deterministic multi-tape 2-way Turing machines are considered. Separation theorems are proved for all constructible space functions of Monte Carlo multi-tape 2-way Turing machines. For instance, palindromes cannot be recognized by these machines in space less than log n. Another example is a language recognized by these machines in space log-star (n) but not in space inverse Ackermann function. Similar separation theorems are proved for all constructible time functions of n + f(n) type for Monte Carlo multi-tape 2-way Turing machines. In these theorems f(n) is linear or sublinear. There is a language recognizable by a deterministic machine in (1+e)n time but not recognizable by a Monte Carlo multi-tape 2-way Turing machine in n+ o(n) time. If g(n) is o(h(n)) then there is a language recognizable by a deterministic machine in n+h(n) time but not recognizable by a Monte Carlo machine in n+g(n) time. Host: Prof. Leonid Levin ------------------------------------------------------------------------------- For colloquium info, including directions, see http://cs-www.bu.edu/colloquium For more information contact Prof. Mark Crovella -------------------------------------------------------------------------------