ADAPTIVE MACHINE LEARNING IN DELAYED FEEDBACK DOMAINS BY SELECTIVE RELEARNING
نویسندگان
چکیده
منابع مشابه
Adaptive Machine Learning in Delayed Feedback Domains by Selective Relearning
We present a novel hybrid technique for improving the predictive performance of an online Machine Learning system: Combining advantages from both memory based and concept based procedures Selective Relearning tackles the problem of learning in gradually changing domains with delayed feedback. The idea is based on training and retraining the model only on the subsegment of the historical dataset...
متن کاملAdaptive tuning of feedback gain in time-delayed feedback control.
We demonstrate that time-delayed feedback control can be improved by adaptively tuning the feedback gain. This adaptive controller is applied to the stabilization of an unstable fixed point and an unstable periodic orbit embedded in a chaotic attractor. The adaptation algorithm is constructed using the speed-gradient method of control theory. Our computer simulations show that the adaptation al...
متن کاملOnline Learning under Delayed Feedback
Online learning with delayed feedback has received increasing attention recently due to its several applications in distributed, web-based learning problems. In this paper we provide a systematic study of the topic, and analyze the effect of delay on the regret of online learning algorithms. Somewhat surprisingly, it turns out that delay increases the regret in a multiplicative way in adversari...
متن کاملLearning in Continuous Domains with Delayed Rewards
Much has been done to develop learning techniques for delayed reward problems in worlds where the actions and/or states are approximated by discrete representations. Although this is acceptable in some applications there are many more situations where such an approximation is diicult and unnatural. For instance, in applications such as robotic,s where real machines interact with the real world,...
متن کاملAutonomous learning by simple dynamical systems with delayed feedback.
A general scheme for the construction of dynamical systems able to learn generation of the desired kinds of dynamics through adjustment of their internal structure is proposed. The scheme involves intrinsic time-delayed feedback to steer the dynamics towards the target performance. As an example, a system of coupled phase oscillators, which can, by changing the weights of connections between it...
متن کاملذخیره 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: Applied Artificial Intelligence
سال: 2008
ISSN: 0883-9514,1087-6545
DOI: 10.1080/08839510802226793