Sign inSign in
ConceptAI & ML

Supervised learning

Learning to predict an output y from an input x given a data set of input-output pairs.
as Sergey Levine defines itUC BerkeleyThe wording closest to all 16 definitions
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

Machine learning4Labeled data3Training data3Target function2Artificial neural networks1Basic premise of learning1Basic probability1Data1

Opens up

Concepts that name this one as a prerequisite.

Lecturers

13

Each in their own terms, from the easiest level.

  • Yaser Abu-Mostafa

    Caltech · Introductory

    A setting where we have a target function and a dataset containing both the input features and the corresponding target outputs.
    • 38:52The Learning Problem
    • 3:32Is Learning Feasible?
  • Fei-Fei Li, Justin Johnson & Serena Yeung

    Stanford University · Undergraduate

    A type of problem where we have data X and labels Y, and the goal is to learn a function mapping X to Y.
    • 1:11Generative Models
    • 1:20Deep Reinforcement Learning
  • Brian Yu

    Harvard University · Undergraduate

    A type of task where we give the computer access to a data set consisting of input-output pairs, and the AI learns a function to map inputs to outputs.
    • 7:20:20Harvard CS50’s Artificial Intelligence with Python – Full University Course
  • Andrew Ng

    Stanford University · Undergraduate

    A learning problem where the algorithm is provided with a data set of inputs and their corresponding right answers (outputs) to learn the association between them.
    • 38:31Machine Learning (CS 229): Lecture 1 - Course Introduction
  • Andrej Karpathy

    Undergraduate

    A training process where models imitate human expert players.
    • 2:43:16Deep Dive into Large Language Models: Pretraining to RLHF
  • Emma Brunskill

    Stanford University · Undergraduate

    A machine learning paradigm that involves learning from data with correct labels and performing generalization.
    • 23:08Introduction to Reinforcement Learning (CS234, 2024, Lecture 1)
  • Shrikant I. Bangdiwala

    Christian Medical College Vellore · Undergraduate

    Training a model on labeled data, where the desired output is known.
    • 9:27Day 4 (IV) - AI, ML and Computer intensive methods in RCTs _ Prof. Shrikant I Bangdiwala
  • Barbara Kwiecinska

    University of Louisville · Undergraduate

    A learning method where data sets with a known output are presented to the neural network to calculate and minimize error.
    • 7:29Current & Future Application of AI in Epilepsy by Dr. Barbara Kwiecinska, PGY4 Neurology Resident
  • Yann LeCun

    Association for Computing Machinery (ACM) · Graduate

    Training a machine by showing examples instead of programming it, where parameters are tweaked when the output is wrong.
    • 3:19Objective-Driven AI: World Models that Learn, Remember, Reason, and Plan
    • 0:12The Deep Learning Revolution: The Sequel (Turing Award Lecture)
  • Ashish Vaswani, Lucas Beyer & William Fedus

    Stanford University · Graduate

    Predicting on the same tasks that supervised deep learning or any supervised machine learning method is trying to predict well on.
    • 20:46Self Attention and Non-parametric transformers (NPTs)
  • Chris Ré

    Stanford University · Graduate

    A machine learning paradigm where a model is trained on a dataset of input-output pairs to learn a mapping from inputs to outputs.
    • 0:27Neural Networks I: Feedforward Architectures & Backpropagation
  • Ilya Sutskever

    UC Berkeley · Graduate

    A paradigm in machine learning that gives precise mathematical conditions under which learning must succeed, requiring low training error and a training set larger than the degrees of freedom.
    • 2:46An Observation on Generalization and Unsupervised Compression

Lectures that assume it

44

The first 12 are shown.

Connections

69

The lectures that define it or assume it, and the concepts next to it.

Concepts9

Show 3 more

Lectures60

Show 54 more

13 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.