learning algorithm
Pattern-finding computational method
A learning algorithm is a procedure that improves its performance on a task by analyzing data, rather than following explicit instructions.
Machine learning systems use learning algorithms to recognize patterns, make predictions, or optimize decisions: The learning algorithm got better at detecting spam as it processed more emails.
The term distinguishes adaptive computational methods from traditional programming, where behavior is fixed in advance.
- neural networkInterconnected computational nodes mimicking cognition
- hidden layersInternal neural network stages
- trainingDeveloping abilities through practice
- training dataInput for machine learning
- dataFacts for analysis
model · backpropagation · optimization · weights · prediction · classification · supervised · unsupervised · pattern recognition · iteration · convergence · bias · features · training set · accuracy · overfitting · parameters · inference
- AIAI, chatbot, neural networks, bot, neural network
- systemsecosystems, nervous system, computer system, immune system, circulatory system
- flashcardsflashcards, card, index cards, cards, digital flashcards
- neural computingneural network, perceptron, backpropagation, brain chip