ClaimsYann LeCun

“There have been many proposals to combine neural nets and structured prediction in the past, going back to the early 1990s.”

Yann LeCun, Quora answer 28 July 2016 Yann LeCun Quora Session Overview (KDnuggets) archived copy

Factual AI capabilities

There have been many proposals to combine neural nets and structured prediction, going back to the early 1990s.

Unresolvable The public record could not settle it.

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

The second judge would not resolve this claim; the verdict shown is the first judge's.

What would count as true or false

Judge as of 2016-07-28. True: the literature contains multiple (several or more) published proposals combining neural networks with structured prediction (for example graph transformer networks, hybrid neural net and HMM systems for speech or handwriting, neural CRFs), with at least some dating from the early 1990s or before. Mostly true: there were multiple such proposals, but the earliest date from the mid-1990s rather than the early 1990s, or 'many' slightly overstates a handful. Mostly false: such work existed but was sparse (one or two proposals), or began only well after the early 1990s. False: no such combinations were proposed before the 2010s.

Set before the research began.

Evidence (1)

  1. E1 ISCA Archive - Hybrid neural network/hidden Markov model continuous-speech recognition ISCA Archive (Proc. ICSLP 1992) · 1992

    “In this paper we present a hybrid multilayer perceptron (MLP)/hidden Markov model (HMM) speaker-independent continuous-speech recognition system, in which the advantages of both approaches are combined by using MLPs to estimate the state-dependent observation probabilities of an HMM.”

Reasoning

The only independent evidence, [E1], records one 1992 proposal that combines a neural network with an HMM. That supports work going back to the early 1990s, but it does not show the multiple proposals that a "true" or "mostly true" verdict needs under the criteria.

Cite this claim

Yann LeCun, “There have been many proposals to combine neural nets and structured prediction, going back to the early 1990s.” (said 28 July 2016). Verdict: unresolvable. Reliability API, claim c_1c09fba2d993, method 0.1.0+c876030c, dataset release v1.1.2. https://epistacklabs.com/claims/c_1c09fba2d993

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 18 March 2024
    “For starters, transformers are equivariant to permutations.”

    Transformers are equivariant to permutations, as ConvNets are equivariant to translations.

  2. Mostly true Factual 14 March 2024
    “If it were the case, we would have AI systems that could teach themselves to drive a car in 20 hours of practice, like any 17 year-old.”

    There are currently no AI systems that can teach themselves to drive a car in about 20 hours of practice, as a 17-year-old can.

  3. Mostly true Factual 8 March 2024
    “The architecture is good, the architecture of the encoder is good, but the fact that you train the system to reconstruct images does not lead it to produce to long, good generic features of images.”

    Training an encoder to reconstruct images does not lead it to produce good generic image features, though the encoder architecture is good.

  4. Mostly true Factual 8 March 2024
    “LLMs can do none of those or they can only do them in a very primitive way and they don’t really understand the physical world.”

    Current LLMs can do none of understanding the physical world, persistent memory, reasoning and planning, or only in a very primitive way.

  5. Mostly true Factual 8 March 2024
    “No, there’s one thing that autoregressive LLMs or that LLMs in general, not just the autoregressive one, but including the bird style bidirectional ones, are exploiting and its self supervised…”

    LLMs, including BERT-style bidirectional ones, exploit self-supervised learning, and LeCun has been a very strong advocate of self-supervised learning for many years.

All of them

Verdicts are made automatically; how verdicts are made.