MINTBIOS / CONNECTED KNOWLEDGE
Deep learning explained
Understand the mechanisms behind the capability.
Build your understanding.
Neural networks explained: inputs, weights and learning
A neural network is a mathematical model that turns numerical inputs into outputs through connected layers of calculations. Training adjusts the numbers inside those calculations so the outputs better match a chosen objective.
EXPLORETraining, inference, retrieval and fine-tuning: which do you need?
Training changes a model’s parameters. Inference uses a model to produce an output. Retrieval supplies information for a request, while fine-tuning is additional training of an existing model. These are different ways of addressing different problems.
EXPLOREOverfitting: why a good demo can fail on new examples
Overfitting happens when a model fits the examples used to train it but performs less well on new examples. Generalisation is its ability to work beyond those training examples. A convincing demonstration is not enough to establish that ability.