A Michigan style architecture for learning finite state controllers: a first step

نویسندگان

  • Olivier Sigaud
  • Samuel Landau
چکیده

In this paper, we describe FACS, a new Michigan style architecture able to build Finite-State Automata controllers for agents learning to solve nonMarkov problems. FACS relies on a population of partial automata and implements a Reinforcement Learning algorithm to compute the strength of each automaton and a Genetic Algorithm to select and discover efficient automata. We detail our approach and present very preliminary results.

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

Implementation of Low-Cost Architecture for Control an Active Front End Rectifier

In AC-DC power conversion, active front end rectifiers offer several advantages over diode rectifiers such as bidirectional power flow capability, sinusoidal input currents and controllable power factor. A digital finite control set model predictive controller based on fixed-point computations of an active front end rectifier with unity displacement of input voltage and current to improve dynam...

متن کامل

Genetic Feedforward-Feedback Controller for Functional Electrical Stimulation Control of Elbow Joint Angle

Background: Functional electrical stimulation (FES) is the most commonly used system for restoring functions after spinal cord injury (SCI). Objective: In this study we investigated feedback PID and feedforward-feedback P-PID controllers for regulating the elbow joint angle. Methods: The controllers were tuned based on a nonlinear muculoskeletal model containing two links, one joint with one de...

متن کامل

Proteome Analysis of Rat Hippocampus Following Morphine-induced Amnesia and State-dependent Learning

Morphine’s effects on learning and memory processes are well known to depend on synaptic plasticity in the hippocampus. Whereas the role of the hippocampus in morphine-induced amnesia and state-dependent learning is established, the biochemical and molecular mechanisms underlying these processes are poorly understood. The present study intended to investigate whether administration of morphine ...

متن کامل

Proteome Analysis of Rat Hippocampus Following Morphine-induced Amnesia and State-dependent Learning

Morphine’s effects on learning and memory processes are well known to depend on synaptic plasticity in the hippocampus. Whereas the role of the hippocampus in morphine-induced amnesia and state-dependent learning is established, the biochemical and molecular mechanisms underlying these processes are poorly understood. The present study intended to investigate whether administration of morphine ...

متن کامل

Learning Document Image Features With SqueezeNet Convolutional Neural Network

The classification of various document images is considered an important step towards building a modern digital library or office automation system. Convolutional Neural Network (CNN) classifiers trained with backpropagation are considered to be the current state of the art model for this task. However, there are two major drawbacks for these classifiers: the huge computational power demand for...

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 2004