#1




Overfitting with Polynomials : deterministic noise
I have some difficulty to understand the idea of deterministic noise. I think there are some disturbing contradiction with what we've seen with the biasvariance tradeoff, particularly with 50th order noiseless target exemple.
Chapter 4 state that:  A 2nd order polynomial could be better than a 10th order polynomial to fit a 50th order polynomial target and it's due to the deterministic noise. > So I conclude that there is more deterministic noise with 10th order than with the 2nd order.  Deterministic noise=bias But we've seen with the biasvariance tradeoff, that a more complex model than an other have a lower bias. Obvioulsy, I'm wrong somewhere, but where ? 
#2




Re: Overfitting with Polynomials : deterministic noise
Quote:
If you want to isolate the impact of deterministic noise on overfitting without interference from the model complexity, you can fix the model and change the complexity of the target function.
__________________
Where everyone thinks alike, no one thinks very much 
#3




Re: Overfitting with Polynomials : deterministic noise
I think the confusion comes from Figure 4.4 compared to the figures of the stochastic noise.
Here you write the shading is the deterministic noise, since this is the difference between the best fit of the current model and the target function. Exactly this shading is from the biasvariance analyses. Thus the value of the deterministic noise is directly related to the bias. When you talk about stochastic noise you say that the outofsample error will increase with the modelcomplexity and this is related to the area between the final hypothesis and the target . Thus the reader might think the bias is increasing with the complexity of the model. However the bias depends on and not on . And the reason why this area increases is due to the stochastic noise. If there isn't any noise the final hypothesis will have a better chance to fit (depending on the position of the samples). In fact (and this is not really clear form the text, but from Exercise 4.3) on a noiseless target the shaded area in Figure 4.4 will decrease when the model complexity increases and thus the bias decreases. My suggestion is to make is more clear, that in case of stochastic noise you talk about the actual final hypothesis and in case of deterministic noise you talk about the best fitting hypothesis, that is related to . From my understanding I would say: Overfitting does not apply to the best fit of the model () but to the real hypothesis (). In the biasvarianceanalyses we saw the variance will increase together with the model complexity (at the same number of samples). So I think Overfitting is a major part of the variance, either due to the stochastic noise or due to the deterministic noise. 
Thread Tools  
Display Modes  

