There are three popular methods for constructing highly retargetable compilers: (1) the compiler emits abstract machine code which is interpreted at run-time, (2) the compiler emits C code which is subsequently compiled to machine code by the native C compiler, or (3) the compiler

There are three popular methods for constructing highly retargetable compilers: (1) the compiler emits abstract machine code which is interpreted at run-time, (2) the compiler emits C code which is subsequently compiled to machine code by the native C compiler, or (3) the compiler

Therearethreepopularmethodsforconstructinghighly retargetablecompilers:(1)thecompileremitsabstractmachinecodewhichisinterpretedatrun-time,(2)thecompiler emitsCcodewhichissubsequentlycompiledtomachine codebythenativeCcompiler,or(3)thecompiler

that the probability of a mutation on an edge e of the tree, conditional on there having been any given number of mutations earlier in the tree, lies strictly between a and b, where 0 Ͻ a Յ b Ͻ 1. We will also suppose that, conditional on (1) a mutation occurring on edge e ϭ (u, v) and (2) given the state at u

mcmxciv This thesis deals with combinatorics in connection with Coxeter groups, finitely generated but not necessarily finite. The representation theory of groups as nonsingular matrices over a field is of immense theoretical importance, but also basic for computational group theory, where the group elements are data structures in a computer. Matrices are unnecessarily large structures, and part of this thesis is concerned with small and efficient representations of a large class of Coxeter groups (including most Coxeter groups that anyone ever payed any attention to.) The main contents of the thesis can be summarized as follows. • We prove that for all Coxeter graphs constructed from an n-path of unlabelled edges by adding a new labelled edge and a new vertex (sometimes two new edges and vertices), there is a permutational representation of the corresponding group. Group elements correspond to integer n-sequences and the nodes in the path generate all n! permutations. The extra node has a more complicated action, adding a certain quantity to some of the numbers.

Commutative, idempotent groupoids and the

Fuzzy logic has been successfully applied in various fields. However, as fuzzy controllers increase in size and complexity, the number of control rules increases exponentially and real-time behavior becomes more difficult. This thesis introduces an any-time fuzzy controller. Much work has been done to optimize and speed up a controlling process, however none of the existing solutions provides an any-time behavior. This study first introduces several constraints that should be satisfied in order to guarantee an any-time behavior. These constraints are related to aggregation and defuzzification phases of fuzzy control. Popular aggregation and defuzzification methods (max-min, sum-product, MOM and COG) are first shown to satisfy these constraints, and then three linearization methods are presented. Linearization methods are used to reorder fuzzy rules base such that a reordered rule base would result in any-time behavior. Finally, several approximation methods are described, that do not break any-time behavior, while causing the intermediate result of an any-time controller to come closer to the final (full calculation) result in a shorter time. The exact influence and worthiness of approximation

This is a proceedings article reviewing a recent combinatorial construction of the bsu(n)k WZNW fusion ring by C. Stroppel and the author. It contains one novel aspect: the explicit derivation of an algorithm for the computation of fusion coecients dierent from the Kac-Walton formula. The discussion is presented from the point of view of a vertex model in statistical mechanics whose partition function generates the fusion coecients. The statistical model can be shown to be integrable by linking its transfer matrix to a particular solution of the Yang-Baxter equation. This transfer matrix can be identied with the generating function of an (innite) set of polynomials in a noncommutative alphabet: the generators of the local ane plactic algebra. The latter is a generalisation of the plactic algebra occurring in the context of the Robinson-Schensted correspondence. One can dene analogues of Schur polynomials in this noncommutative alphabet which become identical to the fusion matrices when represented as endomorphisms over the state space of the integrable model. Crucial is the construction of an eigenbasis, the Bethe vectors, which are the idempotents of the fusion algebra. x

Abstract not found

The compressive sensing (CS) framework aims to ease the burden on analog-to-digital converters (ADCs) by exploiting inherent structure in natural and man-made signals. It has been demon-strated that structured signals can be acquired with just a small number of linear measurements, on the order of the signal complexity. In practice, this enables lower sampling rates that can be more easily achieved by current hardware designs. The primary bottleneck that limits ADC sam-pling rates is quantization, i.e., higher bit-depths impose lower sampling rates. Thus, the decreased sampling rates of CS ADCs accommodate the otherwise limiting quantizer of conventional ADCs. In this thesis, we consider a different approach to CS ADC by shifting towards lower quantizer bit-depths rather than lower sampling rates. We explore the extreme case where each measurement is quantized to just one bit, representing its sign. We develop a new theoretical framework to analyze this extreme case and develop new algorithms for signal reconstruction from such coarsely quantized measurements. The 1-bit CS framework leads us to scenarios where it may be more appropriate to reduce bit-depth instead of sampling rate. We find that there exist two distinct regimes of operation that correspond to high/low signal-to-noise ratio (SNR). In the measurement