DEV Community

24P-0578 Fatima Zoalfqar
24P-0578 Fatima Zoalfqar

Posted on

Understanding Modern AI: Adaptive A* Search and Agentic Ai

Introduction:
Artificial Intelligence is developing very quickly. Researchers are constantly improving AI algorithms and systems. In our AI course we studied topics like search algorithms and intelligent agents.

In this blog, I explored two research papers related to these topics. The first paper improves the A* search algorithm and the second paper explains the concept of Agentic AI.

**

Paper 1: Adaptive A* Search Algorithm:

**

The A* algorithm is one of the most common search algorithms used in artificial intelligence. It is used to find the best or shortest path between two points.

The algorithm uses two values:
the actual cost from the start node and the estimated distance to the goal.

Sometimes the algorithm explores many unnecessary paths which can make the search slower.

The research paper “Research on the A Algorithm Based on Adaptive Weights (2025)”* improves the algorithm by introducing adaptive weights. This means the importance of the heuristic value can change during the search process.

Because of this improvement, the algorithm can focus on better paths and reduce unnecessary exploration. This makes the search faster and more efficient.

This research connects to the informed search algorithms we studied in our AI course.

Paper 2: The Rise of Agentic AI:

The second paper explains the idea of Agentic AI.

Agentic AI systems are artificial intelligence systems that can act more independently. They can observe their environment, make decisions, and perform tasks to achieve goals.

Traditional AI systems usually perform tasks when given instructions. However, agentic AI systems behave more like intelligent agents that can plan actions and solve problems on their own.

The paper also discusses challenges such as safety, reliability, and controlling autonomous AI systems.

This concept connects to the intelligent agents model we studied in class.

Personal Insights

While reading the papers and using NotebookLM, I understood the ideas more clearly. NotebookLM helped explain difficult parts of the research papers in simple language.

One interesting thing I learned is that AI research is improving both search algorithms and intelligent agents. These improvements can help AI systems become faster and more independent.

Video Explanation

Here is my video explaining these papers:

https://youtu.be/SZIPveQmd9g

About Me

Hello! I’m Fatima Zolfqar, a FAST University student interested in Artificial Intelligence.
@raqeeb_26

Top comments (1)

Collapse
 
raqeeb_26 profile image
Raqeeb

Nice work, appreciated.It would be better if you give a complete example where we can use this algorithm in practical life and how it improves our productivity.