Global Optimisation of Neural Network Models via Sequential Sampling
نویسندگان
چکیده
We propose a novel strategy for training neural networks using sequential sampling-importance resampling algorithms. This global optimisation strategy allows us to learn the probability distribution of the network weights in a sequential framework. It is well suited to applications involving on-line, nonlinear, non-Gaussian or non-stationary signal processing.
منابع مشابه
Global optimisation of neural network models via sequential sampling-importance resampling
We propose a novel strategy for training neural networks using sequential Monte Carlo algorithms. This global optimisation strategy allows us to learn the probability distribution of the network weights in a sequential framework. It is well suited to applications involving on-line, nonlinear or non-stationary signal processing. We show how the new algorithms can outperform extended Kalman lter ...
متن کاملHybrid sequential Monte Carlo/Kalman methods to train neural networks in non-stationary environments
In this paper, we propose a novel sequential algorithm for training neural networks in non-stationary environments. The approach is based on a Monte Carlo method known as the sampling-importance resampling simulation algorithm. We derive our algorithm using a Bayesian framework, which allows us to learn the probability density functions of the network weights and outputs. Consequently, it is po...
متن کاملAsynchronous Parallel Bayesian Optimisation via Thompson Sampling
We design and analyse variations of the classical Thompson sampling (TS) procedure for Bayesian optimisation (BO) in settings where function evaluations are expensive, but can be performed in parallel. Our theoretical analysis shows that a direct application of the sequential Thompson sampling algorithm in either synchronous or asynchronous parallel settings yields a surprisingly powerful resul...
متن کاملComputer Assisted Composition with Recurrent Neural Networks
Sequence modeling with neural networks has lead to powerful models of symbolic music data. We address the problem of exploiting these models to reach creative musical goals. To this end we generalise previous work, which sampled Markovian sequence models under the constraint that the sequence belong to the language of a given finite state machine. We consider more expressive non-Markov models, ...
متن کاملPrediction of the Effect of Polymer Membrane Composition in a Dry Air Humidification Process via Neural Network Modeling
Utilization of membrane humidifiers is one of the methods commonly used to humidify reactant gases in polymer electrolyte membrane fuel cells (PEMFC). In this study, polymeric porous membranes with different compositions were prepared to be used in a membrane humidifier module and were employed in a humidification test. Three different neural network models were developed to investigate several...
متن کامل