search

yousuketakada / prml_errata

More PRML Errata
https://yousuketakada.github.io/prml_errata/
Other
80 stars 17 forks source link

Errata for Figures 5.11 and 7.1 #16

Open tommyod opened 5 years ago

tommyod commented 5 years ago

Found two more errata in figures. As always, feel free to edit the text before merging.

Also, on page 343 I believe there is a contradiction. Last sentence in the second to last paragraph states that "at most vN data points fall outside the insensitive tube", and that "at least vN data points lie either on the tube or outside it".

yousuketakada commented 5 years ago

Thank you for your errata.

As for the one for Figure 5.11 on Page 260, I don't see the error you have pointed out. The all sub-figure titles are already in the form "\alpha_1^w = ..., \alpha_1^b = ..., \alpha_2^w = ..., \alpha_2^b = ..."

I am aware that the official errata document (for the first and the second printings) states that all the right hand sides should be raised to -2 (10, 100, 1000 are actually variances); and also that, in the third printing (the one you have), this error seems to have been corrected. I guess there may be again some new errors that have been introduced by the correction. Could you double-check?

tommyod commented 5 years ago

Page 260: In my book (third printing, it says "9 8 (corrected at 8th printing 2009)" in the index page) the sub-figure titles say "\alpha_1^w = ..., \alpha_1^b = ..., \alpha_1^w = ..., \alpha_1^b = ...". Raising the numbers to the power of -2 is indeed corrected in my book. My guess is that correcting one error might have introduced another, since in an earlier printing the indices are correct.

yousuketakada commented 5 years ago

Thank you for your confirmation. I think we have found another regression error in Figure 5.11.

As for Figure 7.1, I agree with you in that we should take the notation of the left sub-figure.