Scribd
Upload
7 views6 pages

Calculus Report

The document discusses the shortest path problem in graph theory, emphasizing its significance in various real-world applications and the A* algorithm, which combines Dijkstra’s method with heuristics for efficient pathfinding. It outlines the strengths and limitations of A*, the concept of dual problems in optimization, and the importance of undominated subsets in improving computational efficiency. Additionally, it provides references for further learning on related algorithms and concepts.

Uploaded by

thinh109.206
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
7 views6 pages

Calculus Report

The document discusses the shortest path problem in graph theory, emphasizing its significance in various real-world applications and the A* algorithm, which combines Dijkstra’s method with heuristics for efficient pathfinding. It outlines the strengths and limitations of A*, the concept of dual problems in optimization, and the importance of undominated subsets in improving computational efficiency. Additionally, it provides references for further learning on related algorithms and concepts.

Uploaded by

thinh109.206
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

I/Definition and topic introduction:

1)Overall: The shortest path problem is a fundamental concept in


graph theory and computer science, focusing on finding the minimum
distance or cost between two points (nodes) in a network. It plays a
critical role in various domains, such as transportation, communication,
and optimization.

2)Graph components and goal: In essence, a graph consists of nodes


connected by edges, each associated with a weight representing
distance, cost, or time. The goal is to determine the optimal path
between a starting node and a target node. Depending on the graph's
characteristics—such as edge weights being non-negative, negative, or
uniform—different algorithms.

3)Applications in reality: Real-world applications of the shortest path


problem are vast, spanning GPS navigation, internet packet routing, and
supply chain management. This topic not only enhances theoretical
understanding of graph algorithms but also contributes to solving
complex, real-world challenges efficiently.

II/A* (A-star) graph:

1)Definition: A* (A-star) is a pathfinding and graph traversal algorithm


used to find the shortest path between two nodes. It combines the
advantages of Dijkstra’s algorithm and a heuristic-based approach to
optimize efficiency.

2)Key features:
+Heuristic Function: Uses a heuristic h(x) to estimate the cost
from a node to the goal. Common heuristics include Manhattan and
Euclidean distances. A* also uses a priority queue to find the shortest
path.
+Priority Queue: Maintains a list of nodes to explore, sorted by
their total estimated cost f(x)=g(x)+h(x), where g(x) is the actual cost to
the node, and h(x) is the heuristic.
+Optimality: A* guarantees the shortest path if h(x) is admissible
(never overestimates the true cost).
+Determining based on positive edge lengths and random costs
(the amount of length).

3)Strengths:
+Highly efficient for large, weighted graphs.
+Adaptive for real-world problems like robotics, video game AI,
and GPS navigation.

4)Limitations:
+Performance depends on heuristic quality.
+Can consume significant memory for dense graphs.

5)Improving the accuracy and speed of function:


+Choosing a suitable Heuristic function.
+Changing function based on different kinds of condition.
+Pruning Heuristic function.

6)Graph and codes: (Nguyên điền vào đây)

III/Theorems and fixed condition:


1)Theorem: Considering the case where w(e), c(e), e ∈ En are
independent random variables and we let Ln = Ln(c0) denote the
random minimum length of a within budget shortest path. Also, let Hn
denote the hop-count (number of edges) in the shortest such path. In
particular we will assume that w(e), c(e) are independent copies of the

Suppose that w(e), c(e), e ∈ En are independent copies


uniform [0, 1] random variable U.

of U.W.h.p (With high percentage) for every path P ∈ P, we have:

2)Condition for expenditure: With every c(e) < c(0); c(0) can be seen
as constrained cost and determined by: c0 ≈ logn/n. (c(n) can be
considered as expenditure or the amount of edge)

IV/Dual problem:

1) Overview: Dual problem is derived from a given primal problem


and provides a different perspective on the same optimization task. The
dual problem often simplifies calculations, offers bounds on the primal
solution, and helps in analyzing the problem's properties. It arises from
the Lagrangian formulation of the primal problem, focusing on
constraints and their associated multipliers.

2)Target:
+Finding the lower bound for optimal value.
+Providing an alternative solution in case of graph being
complicated and including many lengths to minimize calculating time.
3)Duality gap:
Dualtity gap = p*-d*, o(1) = | p*-d*|.
.p* is optimal value of primary calculation.
.d* is optimal value of dual problem.

4)Lagrangian dual: d*=

5)Application in A* graph:
+ Desinging a better heuristic balancing between speed and
accuracy by using λ in the function. For example: h(n)= λ.estimate(n).
+Navigating and considering realistic optimal values. Some issues
in A* graph is not to mention length with strange length or to
underestimate expenditure c(e) in some parts. Mentioned problems
will be fixed properly by using duality.
+Reducing the complication of the graph and optimizing
calculation.

V/Space and codes:

1)Theorem: Let S1, S2, . . . , Sm be a fixed but arbitrary sequence of


subsets of [N]. Suppose that profits are chosen according to the
uniform distribution over [0, 1]. Let q denote the number of dominating
sets over S1, S2, . . . , Sm. Then E(q) = O(N^3).

2)Explanation:The theorem shows the amount of undominated


subset following o(N^3), when N is the amount of lenght in the graph.
Undominated subsets are optimal, smallest and independent subsets.

3)Target:
+Improving the effeciency of the path by declining space
because the theorem only discuss undominated subsets.
+Predicting and controlling the growth of space in the graph A*.
-Code về định lý 11 (cách mà code đưa được định lý này vào rút gọn ko
gian):

VI/References:
1. GeeksforGeeks: Comprehensive tutorials on algorithms like
Dijkstra, Bellman-Ford, Floyd-Warshall, and Johnson’s Algorithm.
Great for both beginners and advanced learners. Visit here
W3Schools.com

Brilliant | Learn interactively


.
2. MIT OpenCourseWare: Lecture notes on all-pairs shortest path
algorithms, including matrix multiplication approaches and
advanced methods like Johnson’s algorithm. Explore here
MIT OpenCourseWare
.
3. HackerEarth: Focuses on the practical implementation of
Bellman-Ford, Dijkstra, and Floyd-Warshall algorithms with code
examples. Check here
HackerEarth
.
4. Brilliant: Visual explanations and interactive tools to understand
graph cycles and algorithm types (single-source, all-pairs). Learn
here
Brilliant | Learn interactively
.
5. Programiz: Simplified, beginner-friendly guide on Dijkstra’s
algorithm with step-by-step implementation. View here
W3Schools.com
.

You might also like