Weak Estimator-Based Stochastic Searching on the Line in Dynamic Dual Environments

Area of Science:

• Computer Science
• Machine Learning
• Artificial Intelligence

Background:

• Stochastic point location involves a learning mechanism (LM) navigating a stochastic environment (SE) using directional cues.
• SEs are either informative (p > 0.5) or deceptive (p < 0.5), influencing LM's success.
• Existing methods struggle with dynamic dual environments where the SE's nature (informative/deceptive) changes over time.

Purpose of the Study:

• To develop a novel solution for LMs to effectively track targets in dynamic dual environments.
• To address the limitations of existing methods in handling time-varying probabilities of environmental cues.

Main Methods:

• A weak estimator-based adaptive step search algorithm is proposed.
• The algorithm enables the LM to adapt to environmental changes in real-time.

Main Results:

• Experimental results demonstrate the proposed solution's feasibility.
• The solution proves to be efficient in tracking targets within dynamic dual environments.

Conclusions:

• The novel weak estimator-based adaptive step search is effective for LMs in dynamic dual environments.
• This approach offers a significant advancement for stochastic point location problems with changing environmental conditions.