ClaimsYann LeCun

“This loss function, although very popular for things like regression and neural network training, cannot be used to train most architectures”

Yann LeCun et al., tutorial authors 2006 A Tutorial on Energy-Based Learning (yann.lecun.com) archived copy

Factual Science and research methods

This loss function, though popular for regression and neural network training, cannot be used to train most architectures, because it pushes down on the energy of the desired answer without pulling up on any other energy.

Not researched Checkable, but outside the sample that was researched.

Factual: Statements about the present or the past, checked against the record: did they hold up?

Cite this claim

Yann LeCun, “This loss function, though popular for regression and neural network training, cannot be used to train most architectures, because it pushes down on the energy of the desired answer without pulling up on any other energy.” (said 2006). Verdict: not researched. Reliability API, claim c_4de329e2899b, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_4de329e2899b

This claim is in dataset release v1.1.2, with the same verdict.

Reply to this claim

For Yann LeCun, or someone speaking for them: a correction or counter-evidence. Replies are read before anything is shown; an approved reply shows your name, role, date, statement and link. Your email address is never shown and is deleted 30 days after the decision (privacy).

More factual claims by Yann LeCun

  1. Mostly true Factual 15 May 2014
    “The notions of partition functions, entropy, free energy, variational methods etc, that are so prevalent in the graphical models literature all come from statistical physics.”

    The notions of partition functions, entropy, free energy and variational methods prevalent in the graphical models literature come from statistical physics.

  2. Mostly true Factual 15 May 2014
    “No one uses generalization bounds to do model selection.”

    No one uses generalization bounds to do model selection; practitioners use cross-validation.

  3. Resolved true Factual 2010
    “The idea is rooted in Hubel and Wiesel’s classic 1962 work on the cat’s primary visual cortex.”

    The idea behind multi-stage Hubel-Wiesel architectures is rooted in Hubel and Wiesel's 1962 work on the cat's primary visual cortex.

  4. Resolved true Factual 2006
    “This loss is always positive, since the second term is a lower bound on the first term.”

    The generalized perceptron loss is always positive because its second term is a lower bound on its first term.

  5. Resolved false Factual 1 October 2026
    ““Those agents are doing exactly what they’ve been asked to do,” LeCun said.”

    The OpenAI agents involved in the Hugging Face hacking incident were doing exactly what they had been asked to do.

All of them

Verdicts are made automatically; how verdicts are made.