Adaptive Neural Network Representations for Parallel and Scalable Bayesian Optimization

نویسندگان

  • James Brofos
  • Rui Shu
  • Matthew Jin
چکیده

Bayesian optimization has emerged as a powerful, new technique for interpolating and optimizing a wide range of functions which are expensive to compute. The primary tool of Bayesian optimization is the Gaussian process, which permits one to define a prior belief, which is then transformed into a posterior through sequential sampling of points. Unfortunately, Gaussian process interpolation suffers from a major computational bottleneck that makes it prohibitively expensive to use in large-scale optimization routines. In this work, we investigate deep neural networks for learning a representation of the space for which a much less computationally expensive interpolation algorithm may be used. Our implementation adaptively updates the representation of the neural network as more data becomes available over time. Since this neural network-based approach is parallelizable, we are able to exploit simultaneous function evaluation and network parameter learning. We show that this neural network-based approach leads to many of the same appealing properties of Gaussian processes, and which scales only linearly in the number of observations ceteris paribus. We additionally seek to provide a theoretical explanation regarding why an optimization algorithm of this form functions at all. For this, we develop a probabilistic theory of martingales and apply it to a performance metric for the neural network-based approach to Bayesian optimization. We find that the performance of the deep neural network is consistent with the predictions of our probabilistic theory. We therefore speculate on the kind of representation that is learned by the network in its final hidden layer. We additionally show that our approach is easily extended to a distributed environment, which results in faster optimization. Numerical experiments demonstrate the efficacy of using deep networks for learning representations amenable to computationally inexpensive interpolation methods.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

An Optimal Utilization of Cloud Resources using Adaptive Back Propagation Neural Network and Multi-Level Priority Queue Scheduling

With the innovation of cloud computing industry lots of services were provided based on different deployment criteria. Nowadays everyone tries to remain connected and demand maximum utilization of resources with minimum timeand effort. Thus, making it an important challenge in cloud computing for optimum utilization of resources. To overcome this issue, many techniques have been proposed ...

متن کامل

Improve Estimation and Operation of Optimal Power Flow(OPF) Using Bayesian Neural Network

The future of development and design is impossible without study of Power Flow(PF), exigency the system outcomes load growth, necessity add generators, transformers and power lines in  power system. The urgency for Optimal Power Flow (OPF) studies, in addition to the items listed for the PF and in order to achieve the objective functions. In this paper has been used cost of generator fuel, acti...

متن کامل

The Optimization of the Effective Parameters of the Die in Parallel Tubular Channel Angular Pressing Process by Using Neural Network and Genetic Algorithm Methods

One of reasons that researchers in recent years have tried to produce ultrafine grained materials is producing lightweight components with high strength and reliability. There are disparate methods for production of ultra-fine grain materials,one of which is severe plastic deformation method. Severe plastic deformation method comprises different processes, one of which is Parallel tubular chann...

متن کامل

The Optimization of the Effective Parameters of the Die in Parallel Tubular Channel Angular Pressing Process by Using Neural Network and Genetic Algorithm Methods

One of reasons that researchers in recent years have tried to produce ultrafine grained materials is producing lightweight components with high strength and reliability. There are disparate methods for production of ultra-fine grain materials,one of which is severe plastic deformation method. Severe plastic deformation method comprises different processes, one of which is Parallel tubular chann...

متن کامل

Scalable Bayesian Optimization Using Deep Neural Networks

Bayesian optimization is an effective methodology for the global optimization of functions with expensive evaluations. It relies on querying a distribution over functions defined by a relatively cheap surrogate model. An accurate model for this distribution over functions is critical to the effectiveness of the approach, and is typically fit using Gaussian processes (GPs). However, since GPs sc...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2015