Overview of artificial neural networks

Area of Science:

• Computational chemistry
• Machine learning in pharmacology
• Artificial intelligence in drug discovery

Background:

• The human brain's structure inspired artificial neural networks (ANNs).
• Modern drug discovery faces a data explosion, necessitating advanced analytical methods.
• ANNs are versatile tools for uncovering complex relationships in large datasets.

Purpose of the Study:

• To introduce the fundamental concepts of artificial neural networks (ANNs).
• To explain the relevance of ANNs in drug discovery modeling.
• To provide foundational knowledge for understanding advanced ANN applications.

Main Methods:

• Overview of ANN principles and their evolution.
• Comparison of ANNs with traditional regression approaches.
• Discussion of commonly used ANN learning methods and network architectures.

Main Results:

• ANNs can model complex nonlinear relationships, outperforming traditional methods.
• ANNs demonstrate excellent fault tolerance, speed, and scalability.
• The chapter provides a foundational understanding of ANNs for drug discovery.

Conclusions:

• Artificial neural networks (ANNs) are powerful tools for analyzing complex data in drug discovery.
• Understanding basic ANN concepts is crucial for leveraging advanced modeling techniques.
• ANNs offer significant advantages over traditional methods for handling large, complex datasets.