Scribd
Upload
9 views10 pages

Untitled Presentation

Vectors are essential in AI, representing data in a mathematical form that machines can process, including feature, weight, and embedding vectors. They face challenges such as high dimensionality, interpretability, and bias, while being utilized in various operations like dot products and cosine similarity for tasks like measuring similarity and optimizing neural networks. Vectors also serve as dynamic arrays in programming, facilitating efficient data storage and manipulation.

Uploaded by

rawan.n.hashem.eng24
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views10 pages

Untitled Presentation

Vectors are essential in AI, representing data in a mathematical form that machines can process, including feature, weight, and embedding vectors. They face challenges such as high dimensionality, interpretability, and bias, while being utilized in various operations like dot products and cosine similarity for tasks like measuring similarity and optimizing neural networks. Vectors also serve as dynamic arrays in programming, facilitating efficient data storage and manipulation.

Uploaded by

rawan.n.hashem.eng24
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Math

Vectors in AI
Students name: rawan nawfal hashim
Dept: p&o eng.

Seminar Iiin AI, vectors represent data in a form that


machines can process mathematically.

Vectors in ai represents the arrays in the codes of programming a


systematic ai .

Quarter Month Year


Vectors are fundamental to the functioning of nearly every AI system. In AI, vectors
represent data in a form that machines can process mathematically. Whether dealing
with numerical inputs, words, images, or behaviors, AI systems use vectors to
capture and manipulate patterns in data.
2. What Is a Vector in AI?

A vector in AI typically refers to an ordered array of numbers, often representing a point or direction in multi-dimensional space. These
can be:

● Feature vectors – Numeric descriptions of input data

● Weight vectors – Parameters learned by models

● Embedding vectors – Encoded forms of complex data like words or images

6. Challenges and Limitations

● Dimensionality: High-dimensional vectors can cause performance and overfitting issues.

● Interpretability: Vector representations, especially learned ones, are often hard to interpret.

● Bias: Vectors can encode and perpetuate biases in data (especially in NLP embeddings).
🔹 1. Dot Product (Inner Product)

● Formula:

● Use in AI :
● Measure similarity between vectors (e.g. cosine similarity )
● Used in neural work ( e.g. neuron activation)

● Use in AI:

○ Measures similarity between vectors (e.g., in NLP, cosine similarity).


🔹 2. Vector Norms (Magnitude / Length)

● L2 Norm (Euclidean):

● L1 Norm (Manhattan):

● Use in AI:
● Regularization to avoid overfitting
● Normalize input vectors
🔹 3. Cosine Similarity

● Formula:

● Use in AI:
● Text similarity, image retrieval , clustering
● Works well with sparse-high dimensional data
🔹 4. Matrix-Vector Multiplication

● Formula:

● Use in AI :
● Core operation in forward propagation in neural network
● Transform data into different feature spaces
🔹 5. Gradient (Vector of Partial Derivatives)

● Formula:

● Use in AI :
● Optimization of loss function vs gradient descent
● Used backpropagation to update weights in neural network
💾 Vectors as Dynamic Arrays (in data structures)
● In languages like C++, a vector is a dynamic array—a resizable array from the Standard Template
Library (STL).
● Efficient for :

● Storing sequential data

● Auto-resizing during insertions

● Index-based access
Conclusion / summary

Vectors in programing could mean mathematical entities ( like in physics)


And data structure like in dynamic arrays and also high dimensional data like ML.
IN THE END OF THIS SEMINAR I WOULD LIKE TO THANK YOU FOR
WATCHING AND LISTENING .

You might also like