For questions about uncertainty quantification (aka uncertainty estimation) in the context of artificial intelligence, in particular, in the context of Bayesian machine learning.
Questions tagged [uncertainty-quantification]
17 questions
37
votes
6 answers
Why do CNN's sometimes make highly confident mistakes, and how can one combat this problem?
I trained a simple CNN on the MNIST database of handwritten digits to 99% accuracy. I'm feeding in a bunch of handwritten digits, and non-digits from a document.
I want the CNN to report errors, so I set a threshold of 90% certainty below which my…
Alexander Soare
- 1,319
- 2
- 11
- 26
7
votes
1 answer
How does the Dempster-Shafer theory differ from Bayesian reasoning?
How does the Dempster-Shafer theory differ from Bayesian reasoning? How do these two methods handle uncertainty and compute posterior distributions?
rudresh dwivedi
- 171
- 3
5
votes
5 answers
How would AI be able to self-examine?
As I see some cases of machine-learning based artificial intelligence, I often see they make critical mistakes when they face inexperienced situations.
In our case, when we encounter totally new problems, we acknowledge that we are not skilled…
A Cat Named Tiger
- 227
- 2
- 5
3
votes
0 answers
Do we need as much information to know if we can can answer a question as we need to actually answer the question?
I am reading The Book of Why: The New Science of Cause and Effect by Judea Pearl, and in page 12 I see the following diagram.
The box on the right side of box 5 "Can the query be answered?" is located before box 6 and box 9 which are the processes…
Lerner Zhang
- 877
- 1
- 7
- 19
3
votes
0 answers
Why does this formula $\sigma^2 + \frac{1}{T}\sum_{t=1}^Tf^{\hat{W_t}}(x)^Tf^{\hat{W_t}}(x_t)-E(y)^TE(y)$ approximate the variance?
How does:
$$\text{Var}(y) \approx \sigma^2 + \frac{1}{T}\sum_{t=1}^Tf^{\hat{W_t}}(x)^Tf^{\hat{W_t}}(x_t)-E(y)^TE(y)$$
approximate variance?
I'm currently reading What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision, and the…
user8714896
- 717
- 1
- 4
- 21
3
votes
0 answers
How can I use Monte Carlo Dropout in a pre-trained CNN model?
In Monte Carlo Dropout (MCD), I know that I should enable dropout during training and testing, then get multiple predictions for the same input $x$ by performing multiple forward passes with $x$, then, for example, average these predictions.
Let's…
lebebop
- 31
- 2
3
votes
1 answer
Is there any research on models that provide uncertainty estimation?
Is there any research on machine learning models that provide uncertainty estimation?
If I train a denoising autoencoder on words and put through a noised word, I'd like it to return a certainty that it is correct given the distribution of data it…
user8714896
- 717
- 1
- 4
- 21
1
vote
2 answers
How do language models know what they don't know - and report it?
Again and again I ask myself what goes on in a pre-trained transformer-based language model (like ChatGPT9) when it comes to "know" that it cannot give an appropriate answer and either
states it ("I have not enough information to answer this…
Hans-Peter Stricker
- 811
- 1
- 8
- 20
1
vote
0 answers
Active Learning regression with Random Forest
I have a dataset of about 8k points and I am trying to employ active learning with the random forest regressor. I have split the dataset to train and test with train being around 20 points. The test serves as the unlabelled pool (although I have the…
ado sar
- 150
- 4
1
vote
0 answers
How to calculate uncertainty in Deep Ensembles for Reinforcement Learning?
Lets take the following example: I must predict the return (Q-values) of x state-action pairs using an ensemble of m models. Using NumPy I could have the following for x = 5 and m = 3:
>>> predictions = np.random.rand(3, 1, 5)
[[[0.22668968…
HenDoNR
- 81
- 4
1
vote
0 answers
Does MobileNet SSD v2 only capture aleatoric uncertainty (and so not the epistemic one)?
Regarding the MobileNet SSD v2 model, I was wondering to what extend it captures uncertainty of the predictions.
There are 2 types of uncertainty, data uncertainty (aleatoric) and model uncertainty (epistemic).
The model outputs bounding boxes with…
Baka
- 11
- 1
1
vote
2 answers
Why is my Keras prediction always close to 100% for one image class?
I am using Keras (on top of TF 2.3) to train an image classifier. In some cases I have more than two classes, but often there are just two classes (either "good" or "bad"). I am using the tensorflow.keras.applications.VGG16 class as base model with…
Matthias
- 155
- 7
0
votes
0 answers
Do deep ensembles and regular ensembles coincide for classification tasks?
The deep ensemble paper https://arxiv.org/pdf/1612.01474.pdf introduces proper scoring rules for ensembles of NNs. Turns out that the likelihood is always a proper scoring rule.
For regression tasks, we can then use the gaussian NLL, which includes…
0
votes
0 answers
What is the relationship between entropy in thermodynamics and entropy in information theory?
BACKGROUND: In thermodynamics, entropy $S$ is a measure of disorder and is given by
$${\displaystyle S=k_B\log(W)},$$
where $k_B$ is Boltzman's constant and $W$ is the number of microstates.
In information theory, (Shannon) entropy $H$ is a measure…
Snehal Patel
- 912
- 1
- 1
- 25
0
votes
1 answer
What does AUSE metric mean in uncertainty estimation
I am reading the paper "Evaluating Scalable Bayesian Deep Learning Methods for Robust Computer Vision", I do not understand the definition of AUSE metric in this sentence "but only in terms of the AUSE metric which is a relative measure of the…
TimothyShi
- 1
- 1