Perspective view of autonomous control in unknown environment: Dual control for exploitation and exploration vs reinforcement learning

نویسندگان

چکیده

This paper overviews and discusses the relationship between Reinforcement Learning (RL) recently developed Dual Control for Exploitation Exploration (DCEE). It is argued that there are two related but quite distinctive approaches, namely, control machine learning, in tackling intractability arising optimal decision making/control problems. In approach, original problems (of an infinite horizon) approximated by finite horizon solved online taking advantage of availability computing power. learning solutions through iterations, or (offline) training trials when models not available. When dealing with unknown environments, DCEE as a technique from approach could potentially solve similar RL while offering number advantages, most notably, coping uncertainty environment/tasks, high efficiency balancing exploitation exploration, potential establishing its formal properties like stability. The links other relevant methods dual control, Model Predictive particularly Active Inference neuroscience discussed. latter provides strong biological endorsement DCEE. discussions illustrated autonomous source search using robot. concluded promising, complementary to RL, more research required develop it generic theory fully realise potential. relationships revealed this provide insights into these facilitate cross fertilisation developing under uncertain environments.

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

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Control of exploitation-exploration meta-parameter in reinforcement learning

In reinforcement learning (RL), the duality between exploitation and exploration has long been an important issue. This paper presents a new method that controls the balance between exploitation and exploration. Our learning scheme is based on model-based RL, in which the Bayes inference with forgetting effect estimates the state-transition probability of the environment. The balance parameter,...

متن کامل

development and implementation of an optimized control strategy for induction machine in an electric vehicle

in the area of automotive engineering there is a tendency to more electrification of power train. in this work control of an induction machine for the application of electric vehicle is investigated. through the changing operating point of the machine, adapting the rotor magnetization current seems to be useful to increase the machines efficiency. in the literature there are many approaches wh...

15 صفحه اول

Adapting Control Methods for Autonomous Exploration of Unknown Environments

On the other hand, developing and testing Proposed missions to explore comets and moons will encounter environments that are hostile and unpredictable. Any successful explorer must be able to adapt to a wide range of possible operating conditions in order to survive. The traditional approach of constructing special-purpose control methods would require a information about the environment, which...

متن کامل

the relationship between locus of control and iranian efl university students’ beliefs about language learning

this exploratory study aimed to investigate a possible relationship between learners’ beliefs about language learning and one of their personality traits; that is,locus of control (loc). both variables, beliefs and locus of control, are assumed to influence the language learning process. the internal control index (ici) and the beliefs about language learning inventory (balli) were administered...

ذخیره در منابع من


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

ژورنال

عنوان ژورنال: Neurocomputing

سال: 2022

ISSN: ['0925-2312', '1872-8286']

DOI: https://doi.org/10.1016/j.neucom.2022.04.131