Deep learning
Neural networks from the ground up: layers, optimisers, CNNs, RNNs, GANs and training tricks.
Page 5 of 10
- The Math Behind Neural Network Learning Backpropagation7 min
Backpropagation is a fundamental algorithm in training artificial neural networks. It's used to calculate gradients of the loss function with respect to the...
- Understanding Backpropagation in Neural Networks9 min
Backpropagation is a fundamental algorithm in neural networks that calculates gradients of the loss function with respect to network weights. It's the key...
- Understanding Backpropagation with Python4 min
"Decoding Backpropagation: The Engine of Neural Networks"
- Understanding Gradient Descent for AI Model Training14 min
Gradient descent is an iterative optimization algorithm used to minimize a loss function by iteratively adjusting parameters in the direction of steepest...
- Understanding Optimizers in Neural Networks13 min
Gradient descent forms the foundation of optimization in neural networks. It iteratively adjusts parameters by computing gradients of the loss function with...
- Understanding Saddle Points in Deep Learning Optimization9 min
The saddle point phenomenon represents a critical challenge in deep learning optimization where the gradient becomes zero but the point is neither a local...
- Vanishing and Exploding Gradient Problems in Python7 min
Gradient problems are fundamental challenges in training deep neural networks. They occur when gradients become too small (vanishing) or too large...
- Vanishing and Exploding Gradients in Deep Learning with Python10 min
Vanishing and exploding gradients are common problems in training deep neural networks. They occur during backpropagation when gradients become extremely...
- Vanishing vs. Exploding Gradients13 min
A fundamental implementation demonstrating vanishing gradients in deep neural networks using sigmoid activation. The network architecture deliberately uses...
- Advanced Generative Adversarial Networks with Python11 min
Generative Adversarial Networks (GANs) are a class of deep learning models consisting of two neural networks: a generator and a discriminator. These networks...
- Adversarial Attacks on Machine Learning Models in Python7 min
Introduction to Adversarial Attacks on Machine Learning Models
- Adversarial Attacks on Neural Networks12 min
Adversarial attacks are techniques used to manipulate input data, often images, to deceive machine learning models, particularly neural networks. These...
- Applying Geometric Deep Learning with Python7 min
Geometric Deep Learning (GDL) is a rapidly evolving field that applies deep learning techniques to non-Euclidean data structures such as graphs and...
- Behind the Scenes of Generative Adversarial Networks (GANs)10 min
Generative Adversarial Networks (GANs) are a class of machine learning models consisting of two neural networks: a generator and a discriminator. The...
- Building a Generative Neural Network from Scratch in Python9 min
Generative Neural Networks (GNNs) are a class of artificial intelligence models capable of creating new data that resembles the training data. They learn the...
- Building Dynamic Topology Neural Networks with Multi-Objective Optimization in Python10 min
Dynamic Topology Neural Networks (DTNNs) are an advanced form of artificial neural networks that can adapt their structure during training. This adaptation...
- Diffusion Models for Imaging and Vision in Python10 min
Diffusion models are a class of generative models that have gained significant attention in the field of computer vision and image processing. These models...
- Exploring 1x1 Convolutions in Deep Learning9 min
1x1 convolutions, also known as pointwise convolutions, are a special type of convolutional layer in neural networks. They operate on a single pixel across...
- Fractal Patterns in Neural Network Hyperparameter Grids12 min
Neural networks are complex systems with numerous hyperparameters. When performing a grid search over these parameters, unexpected patterns can emerge. This...
- FractalNet An Alternative to Residual Neural Networks Using Python7 min
FractalNet is a novel architecture proposed as an alternative to residual neural networks. It is designed to address the optimization difficulties...
- Generative Adversarial Networks in Python:10 min
Generative Adversarial Networks are a class of deep learning models consisting of two neural networks that compete against each other. The generator network...
- Geometric Measure Theory and Non-Archimedean Neural Networks in Python9 min
Geometric Measure Theory (GMT) is a field of mathematics that extends classical measure theory to geometric settings. It provides tools for analyzing complex...
- Identity Mappings in Deep Residual Networks with Python10 min
Identity mappings are a key component in deep residual networks, enhancing gradient flow and easing optimization. They allow for the creation of very deep...
- Introduction to Diffusion Models in Python9 min
Diffusion models are a class of generative models that learn to gradually denoise data. They start with pure noise and iteratively refine it into...
- Introduction to Padding, Strides, and Pooling in Deep Learning with Python9 min
Padding, strides, and pooling are fundamental concepts in deep learning, particularly in convolutional neural networks (CNNs). These techniques help manage...
- Matrix Multiplication in Deep Learning4 min
Matrix multiplication is a fundamental operation in deep learning, crucial for processing inputs, applying weights, and propagating information through...
- Matrix Orientation, Hidden-state Alignment, and Knowledge Distillation in Python9 min
Matrix orientation is a crucial concept in linear algebra and machine learning. It refers to how matrices are arranged and interpreted, affecting...
- Optimizing Diffusion Models for Faster Generation7 min
The DPM-Solver framework revolutionizes diffusion model sampling by implementing a high-order numerical solver based on differential equations theory. It...
- Posterior Collapse in VAEs vs. Diffusion Models10 min
Posterior collapse is a phenomenon that can occur in Variational Autoencoders (VAEs) but is less common in diffusion models. This presentation will explore...
- Response:11 min
Kolmogorov-Arnold Networks (KANs) are a class of neural networks inspired by the Kolmogorov-Arnold representation theorem. These networks aim to approximate...