Results for "deep learning"
Deep Learning
IntermediateA branch of ML using multi-layer neural networks to learn hierarchical representations, often excelling in vision, speech, and language.
Deep Learning is a type of machine learning that uses structures called neural networks, which are inspired by how the human brain works. Imagine a series of layers where each layer learns to recognize different features of an image, like edges, shapes, and eventually, whole objects. This is how ...
A branch of ML using multi-layer neural networks to learn hierarchical representations, often excelling in vision, speech, and language.
Activation max(0, x); improves gradient flow and training speed in deep nets.
Allows gradients to bypass layers, enabling very deep networks.
Gradients grow too large, causing divergence; mitigated by clipping, normalization, careful init.
Designing input features to expose useful structure (e.g., ratios, lags, aggregations), often crucial outside deep learning.
Deep learning system for protein structure prediction.
Simplified Boltzmann Machine with bipartite structure.
Early architecture using learned gates for skip connections.
A point where gradient is zero but is neither a max nor min; common in deep nets.
Limiting gradient magnitude to prevent exploding gradients.
Automatically learning useful internal features (latent variables) that capture salient structure for downstream tasks.
Gradually increasing learning rate at training start to avoid divergence.
Iterative method that updates parameters in the direction of negative gradient to minimize loss.
Methods like Adam adjusting learning rates dynamically.
Reusing knowledge from a source task/domain to improve learning on a target task/domain, typically via pretrained models.
Strategy mapping states to actions.
Combines value estimation (critic) with policy learning (actor).
AI limited to specific domains.
Expected return of taking action in a state.
Probabilistic energy-based neural network with hidden variables.
Visualization of optimization landscape.
Flat high-dimensional regions slowing training.
A gradient method using random minibatches for efficient training on large datasets.
Popular optimizer combining momentum and per-parameter adaptive step sizes via first/second moment estimates.
Gradients shrink through layers, slowing learning in early layers; mitigated by ReLU, residuals, normalization.
Techniques that stabilize and speed training by normalizing activations; LayerNorm is common in Transformers.
Expanding training data via transformations (flips, noise, paraphrases) to improve robustness.
AI focused on interpreting images/video: classification, detection, segmentation, tracking, and 3D understanding.
Optimization with multiple local minima/saddle points; typical in neural networks.
Variability introduced by minibatch sampling during SGD.