News for hackers. Stuff that matters.
Re:Double Descent in Human Learning by jwilber on Wednesday April 26, 2023 @03:11AM (#35710423)
There’s definitely more to the picture than that - it’s really about how models in the interpolation region behave (see https://mlu-explain.github.io/double-descent/ or https://mlu-explain.github.io/double-descent2/)
Re:Double Descent in Human Learning by esafak on Wednesday April 26, 2023 @03:36AM (#35710555)
Which assumption of the bias-variance decomposition is flawed? I assume the variance term must not be increasing as supposed, because overparameterization enables sparsity?