A simple tuning scheme that shifts the crossover frequency of the start-up IIR half-band filter pair to the desired location has been introduced recently. In this paper, the approximation of the tuning scheme is derived in a manner that makes this scheme suitable for fixedpoint arithmetic. The effects of the approximation error to the frequency responses and to the complementary properties for ...

Latent Gaussian Models (LGM) provide a rich modeling framework with general inference procedures. The variational approximation offers an effective solution for such models and has attracted a significant amount of interest. Recent work proposed a fixedpoint (FP) update procedure to optimize the covariance matrix in the variational solution and demonstrated its efficacy in specific models. The ...

We generalize Abrahamse’s interpolation theorem from the setting of a multiply connected domain to that of a more general Riemann surface. Our main result provides the scalar-valued interpolation theorem for the fixedpoint subalgebra of H∞ associated to the action of a Fuchsian group. We rely on two results from a paper of Forelli. This allows us to prove the interpolation result using duality ...

The model-checking games associated with fixed-point logics are parity games, and it is currently not known whether the strategy problem for parity games can be solved in polynomial time. We study Solitaire-LFP, a fragment of least fixedpoint logic, whose evaluation games are nested soltaire games. This means that on each strongly connected component of the game, only one player can make nontri...

Training neural networks is a timeand compute-intensive operation. This is mainly due to the large amount of floating point tensor operations that are required during training. These constraints limit the scope of design space explorations (in terms of hyperparameter search) for data scientists and researchers. Recent work has explored the possibility of reducing the numerical precision used to...

Echo State Networks (ESN) have demonstrated their efficiency in supervised learning of time series: a ”reservoir” of neurons provide a set of dynamical systems that can be linearly combined to match the target dynamics, using a simple quadratic optimisation algorithm to tune the few free parameters. In an unsupervised learning context, however, another optimiser is needed. In this paper, an ada...

In this paper, we present a Catapult C/C++ based methodology that integrates key technologies for high-level VLSI modelling of 3G/4G wireless systems to enable extensive time/area tradeoff study. A Catapult C/C++ based architecture scheduler transfers the major workload to the algorithmic C/C++ fixedpoint design. Prototyping experiences are presented to explore the VLSI design space extensively...

Recent studies have demonstrated that correntropy is an efficient tool for analyzing higher-order statistical moments in nonGaussian noise environments. Although correntropy has been used with complex data, no theoretical study was pursued to elucidate its properties, nor how to best use it for optimization . This paper presents a probabilistic interpretation for correntropy using complex-value...

We study the Newton method and a fixed-point method for solving the system of nonlinear equations arising from the Sinc-Galerkin discretization of the Burgers equations. In each step of the Newton method or the fixedpoint method, a structured subsystem of linear equations is obtained and needs to be solved numerically. In this paper, preconditioning techniques are applied to solve such linear s...

In Dynamic Programming, convergence of algorithms such as Value Iteration or Policy Iteration results -in discounted problemsfrom a contraction property of the back-up operator, guaranteeing convergence to its fixedpoint. When approximation is considered, known results in Approximate Policy Iteration provide bounds on the closeness to optimality of the approximate value function obtained by suc...