Close Menu
    Trending
    • Scaling Vector Search: Comparing Quantization and Matryoshka Embeddings for 80% Cost Reduction
    • Pragmatic by design: Engineering AI for the real world
    • I Finally Built My First AI App (And It Wasn’t What I Expected)
    • Are OpenAI and Google intentionally downgrading their models?
    • 3 Questions: On the future of AI and the mathematical and physical sciences | MIT News
    • Is Open AI actually making its own models dumber?
    • An Intuitive Guide to MCMC (Part I): The Metropolis-Hastings Algorithm
    • New MIT class uses anthropology to improve chatbots | MIT News
    ProfitlyAI
    • Home
    • Latest News
    • AI Technology
    • Latest AI Innovations
    • AI Tools & Technologies
    • Artificial Intelligence
    ProfitlyAI
    Home » Why Are Convolutional Neural Networks Great For Images?
    Artificial Intelligence

    Why Are Convolutional Neural Networks Great For Images?

    ProfitlyAIBy ProfitlyAIMay 1, 2025No Comments4 Mins Read
    Share Facebook Twitter Pinterest LinkedIn Tumblr Reddit Telegram Email
    Share
    Facebook Twitter LinkedIn Pinterest Email


     Universal Approximation Theorem states {that a} neural community with a single hidden layer and a nonlinear activation perform can approximate any steady perform. 

    Sensible points apart, such that the variety of neurons on this hidden layer would develop enormously giant, we don’t want different community architectures. A easy feed-forward neural community may do the trick.

    It’s difficult to estimate what number of community architectures have been developed. 

    Whenever you open the favored AI mannequin platform Hugging Face right this moment, one can find a couple of million pretrained fashions. Relying on the duty, you’ll use completely different architectures, for instance transformers for pure language processing and convolutional networks for picture classification.

    So, why will we want so many Neural Network architectures?

    On this put up, I would like provide a solution to this query from a physics perspective. It’s the construction within the information that conjures up novel neural community architectures.

    Symmetry and invariance

    Physicists love symmetry. The basic legal guidelines of physics make use of symmetries, reminiscent of the truth that the movement of a particle could be described by the identical equations, no matter the place it finds itself in time and house.

    Symmetry at all times implies invariance with respect to some transformation. These ice crystals are an instance of translational invariance. The smaller buildings look the identical, no matter the place they seem within the bigger context.

    By Picture by PtrQs, CC BY-SA 4.0, https://commons.wikimedia.org/w/index.php?curid=127396876

    Exploiting symmetries: convolutional neural community

    In the event you already know {that a} sure symmetry persists in your information, you’ll be able to exploit this reality to simplify your neural community structure.

    Let’s clarify this with the instance of picture classification. The panel reveals three scenes together with a goldfish. It could possibly present up in any location inside the picture, however the picture ought to at all times be labeled as goldfish.

    Three panels showing the same goldfish in different locations
    Pictures created by the writer utilizing Midjourney.

    A feed-forward neural community may definitely obtain this, given enough coaching information. 

    This community structure requires a flattened enter picture. Weights are then assigned between every enter layer neuron (representing one pixel within the picture) and every hidden layer neuron. Additionally, weights are assigned between every neuron within the hidden and the output layer.

    Together with this structure, the panel reveals a “flattened” model of the three goldfish photos from above. Do they nonetheless look alike to you?

    Schematic explaining the flattening of an image and the resulting network architecture.
    Picture created by the writer. Utilizing photos created by the writer with Midjourney and ANN structure created with https://alexlenail.me/NN-SVG/.

    By flattening the picture, now we have incurred two issues:

    • Pictures that include an analogous object don’t look alike as soon as they’re flattened,
    • For prime-resolution photos, we might want to practice plenty of weights connecting the enter layer and the hidden layer.

    Convolutional networks, then again, work with kernels. Kernel sizes sometimes vary between 3 and seven pixels, and the kernel parameters are learnable in coaching.

    The kernel is utilized like a raster to the picture. A convolutional layer could have a couple of kernel, permitting every kernel to give attention to completely different points of the picture. 

    Picture created by the writer.

    For instance, one kernel would possibly decide up on horizontal traces within the picture, whereas one other would possibly decide up on convex curves.

    Convolutional neural networks protect the order of pixels and are nice to study localized buildings. The convolutional layers could be nested to create deep layers. Together with pooling layers, high-level options could be discovered.

    The ensuing networks are significantly smaller than in case you would use a fully-connected neural community. A convolutional layer solely requires kernel_size x kernel_size x n_kernel trainable parameters. 

    It can save you reminiscence and computational price range, all by exploiting the truth that your object could also be situated wherever inside your picture!

    Extra superior deep studying architectures that exploit symmetries are Graph Neural Networks and physics-informed neural networks.

    Abstract

    Convolutional neural networks work nice with photos as a result of they protect the native data in your picture. As a substitute of flattening all of the pixels, rendering the picture meaningless, kernels with learnable parameters decide up on native options.


    Additional studying



    Source link

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleModern GUI Applications for Computer Vision in Python
    Next Article Beyond Glorified Curve Fitting: Exploring the Probabilistic Foundations of Machine Learning
    ProfitlyAI
    • Website

    Related Posts

    Artificial Intelligence

    Scaling Vector Search: Comparing Quantization and Matryoshka Embeddings for 80% Cost Reduction

    March 12, 2026
    Artificial Intelligence

    I Finally Built My First AI App (And It Wasn’t What I Expected)

    March 12, 2026
    Artificial Intelligence

    3 Questions: On the future of AI and the mathematical and physical sciences | MIT News

    March 11, 2026
    Add A Comment
    Leave A Reply Cancel Reply

    Top Posts

    Unlocking AI’s full potential requires operational excellence

    October 1, 2025

    AI Operations Under the Hood: Challenges and Best Practices

    September 5, 2025

    AI-agenter kan nu hjälpa läkare fatta bättre beslut inom cancervård

    June 7, 2025

    Amazons ”House of David” använde över 350 AI-scener

    November 16, 2025

    What It Can and Can’t Do Today

    February 27, 2026
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    Most Popular

    Odyssey AI-modell förvandlar videor till interaktiva 3D-värld

    June 2, 2025

    How To Choose The Perfect AI Tool In 2025 » Ofemwire

    October 22, 2025

    Microsoft’s Quiet AI Layoffs, US Copyright Office’s Bombshell AI Guidance, 2025 State of Marketing AI Report, and OpenAI Codex

    May 20, 2025
    Our Picks

    Scaling Vector Search: Comparing Quantization and Matryoshka Embeddings for 80% Cost Reduction

    March 12, 2026

    Pragmatic by design: Engineering AI for the real world

    March 12, 2026

    I Finally Built My First AI App (And It Wasn’t What I Expected)

    March 12, 2026
    Categories
    • AI Technology
    • AI Tools & Technologies
    • Artificial Intelligence
    • Latest AI Innovations
    • Latest News
    • Privacy Policy
    • Disclaimer
    • Terms and Conditions
    • About us
    • Contact us
    Copyright © 2025 ProfitlyAI All Rights Reserved.

    Type above and press Enter to search. Press Esc to cancel.