Questions tagged [statistics]
The statistics tag has no usage guidance.
144
questions
1
vote
0
answers
13
views
Why use Welford's Method over a more naive approach?
I came across this pair of blog posts explaining the problem with calculating variance on floating point data:
https://jvns.ca/blog/2023/01/13/examples-of-floating-point-problems/#example-3-a-variance-...
0
votes
0
answers
14
views
What is the advantage of every Bayes Nets Sampling algorithm?
I have studied four Bayes Nets Sampling algorithms for calculating the probabilities, including:
$1. Prior Sampling$
$2. Rejection Sampling$
$3. Likelihood Weighting$
$4. Gibbs Sampling$
And my ...
0
votes
0
answers
26
views
How to create random pseudo samples from a non-stationary time series containing NaNs using Monte-Carlo method?
I am looking for answers creating 500 samples of non-stationary Time series based on its probability distribution preferably using monte-carlo simulation. DATA LINK
Secondly, the most example ...
1
vote
1
answer
56
views
Efficient algorithm for finding linear combination of piecewise functions
While porting an algorithm, I having a bit of a problem with finding an efficient algorithm for finding a linear combination of piecewise functions. The procedure is described in the Table 3 of the ...
1
vote
1
answer
40
views
Naïve array sampling algorithm: the possibility of a item being chosen and its time complexity
This naïve sampling algorithm I am talking about is fairly simple: create a set for storing chosen items first, randomly select an item from the array, and examminate if it is in the set. If it isn't, ...
0
votes
0
answers
28
views
Tight bounds for expected maximum of k binomial(n,p) IIDs
What is the tightest lower and upper bound for the expected maximum value of k IID Binomial(n, p) random variables
I tried to derive it :
$$Pr[max \leq C] = (\sum_{i = 0}^C {n \choose i}p^i(1 - p)^i)^...
1
vote
1
answer
32
views
Distribution maximizing ratio of expected maximum over the mean
I’m looking for a distribution that is non-negative, or has good tail bounds (so non-negative with high probability) and maximizes the ratio between the expected maximum of $n$ iid samples and the ...
2
votes
0
answers
50
views
Algorithm for half-sample mode better than O(N^2)
The half-sample mode is a mode estimator that homes in on the highest density region of a set of samples in search of the mode. It is one of the better mode estimators, though it fails for J-shaped ...
1
vote
0
answers
31
views
Precise definition of Universal Learner in Machine Learning
It is surprising to me that I cannot find a precise definition of universal learner on the internet. I can guess what it should bebut I don't want to make a mistake, therefore I have come here.
Here's ...
1
vote
0
answers
89
views
Hoeffding's inequality applicability for sample complexity
I am trying to determine some bounds for sample complexity. Suppose we have a bounded loss function $\ell$ and target function $f:\mathcal{X}\to\mathcal{Y}$. Hypothesis $h$ is learned, then the ...
1
vote
0
answers
74
views
Computational complexity of median filter in image processing
I found this paper http://nomis80.org/ctmf.pdf, that describes an efficient algorithm for applying a median filter to an image. Though this works for non-HDR images, I am not sure how well it scales ...
4
votes
2
answers
866
views
What empirical evidence do we have for or against a correlation between fault density and LOC?
LOC = lines of code
KLOC = Thousand lines of code
Fault (or defect) density = number of reported bugs per line of code.
Software artifact = function, class, module
Reading research papers on fault ...
0
votes
1
answer
41
views
Efficient random sampling from large discrete distribution
I have a random variable $X$ that can take finite values in $\{X_1, ..., X_n\}$ with probabilities $\{p_1,..., p_n\}$. Is there a computationally efficient way to sample a number from this set? My ...
1
vote
0
answers
13
views
Distributed Graph Consensus to fit a distribution?
$G$ is a strongly connected graph with nodes $V$ and edges $E$. Each node $v_i$ receives a sample $x_i$ from a Gaussian $\mathcal{N}(\mu,\sigma^2)$ with unknown mean and variance.
The objective is for ...
1
vote
0
answers
19
views
Distribution independence property testing
I have been reading the proof in the following paper, and I am unable to understand some parts in the proof. This paper shows that a distribution $A$ over $[n]\times[m]$, $n\geq m$, can be $\epsilon$-...
1
vote
2
answers
241
views
Sliding window max/min algorithm without dynamic allocations
I'm working on a suite of DSP tools in Rust, and one of the features of the library is a collection of windowed statistics (mean, RMS, min, max, median, etc) on streams of floating-point samples. For ...
1
vote
1
answer
50
views
EM algorithm - What happens with the standard deviation?
What have you tried?
So I watched this video.
According to the video, we've to calculate the variance $\sigma^2$ as follows:
$$
\sigma_{k}^{2} =
\frac
{
p_{1} \left(x_{1} -...
0
votes
1
answer
40
views
PCA applicability
I understand that PCA takes a data set with input size, with output labels, and reduces the inputs to a set of principal components size r, where r < n.
My question is whether or not this can be ...
0
votes
0
answers
61
views
Most popular path in weighted cylic directed graph
Context
I have a graph $G=(V,E)$ with weighted edges, all weights are positive integers $w(e)\in\mathbb{N}\setminus\{0\}$. The weights represent the popularity/count of each edge, for example $w(e) = ...
1
vote
1
answer
220
views
Approximate max weight path in directed graph
Context
This question is related to the fact one can't use Bellman-Ford to find max weight paths in directed graphs with cycles. The reason is that giving a new graph $\tilde{G}$ with negative weights ...
0
votes
1
answer
280
views
Chebyshev’s inequality problem in one exercises I can't understand if I did it right or not
This is what do I have to solve:
Byron Book: Exercise 8.3 chapter 8
Verify the use of Chebyshev’s inequality in (8.6) of Example 8.16. Show that if the population
mean is indeed 48.2333 and the ...
2
votes
0
answers
22
views
Separating labelled points with a plane, minimizing label variance
Suppose we have observations with associated labels $\{({\bf x}_1, y_1), ({\bf x}_2, y_2), \dots, ({\bf x}_n, y_n)\}$ where ${\bf x}_i \in \mathbb{R}^d$ and $y_i \in \mathbb{R}$.
Can we efficiently ...
1
vote
0
answers
14
views
Tracking approximate values of specific percentiles
At the moment I use HdrHistogram to track approximate distribution of a stream of values. But I don't really need all percentiles, just three of them. Is there some algorithm to track specific ...
1
vote
0
answers
70
views
Differential Privacy with only positive noise
The Laplace mechanism is a standard way of making the output of a function $f$ differentially private.
More concretely, let $\Delta_f$ be the sensitivity of $f$, i.e. the maximum value by which the ...
0
votes
1
answer
52
views
How to implement conditional probability distribution on set-valued Random Variables
I'm trying to implement conditional probability distribution when the events of two RVs are sets. If I try to extrapolate concepts from real or categorical variables to sets things become confusing ...
2
votes
1
answer
40
views
Averaging a list of variances
In a setting where I have n clients each owns a subset of the dataset, each computes the mean and variance on his data and send them to the server. The server would average all means and variances. ...
2
votes
1
answer
56
views
Is this statistical group summation unambiguously reversible?
Let $X$ be a finite multisubset of $\mathbb{N}^2$. Let's introduce the following notation:
$A$ is a set of all first elements of pairs from $X$ and $B$ is a set of all second elements of pairs from $X$...
2
votes
0
answers
24
views
Algorithm to mapping given probabilities to empirical probabilities
Consider following problem statement:
You have given $n$ actions. You can perform any of them. Each action gives you success with some probability. The challenge is to perform given finite number of ...
1
vote
2
answers
144
views
Find the average number of steps to sort an array by randomly selecting two elements
I have sequence of an unique numbers from 1 to 10 in randomly order (for example: list = [7, 5, 3, 4, 2, 6, 10, 1, 9, 8]). I can choose two random number and if the list from left number larger then ...
1
vote
1
answer
22
views
SGD statistical guarantee
I have a question regard online learning with SGD. Is there a way to give a statistical guarantee that the value obtained after $n$ samples deviates at most $\epsilon$ from the real value?
Thank you ...
0
votes
0
answers
34
views
How to handle distribution of values with same attributes into different classes
I'm a student studying a data mining course and have come across a problem.
I need to explain the problem with the help of an example scenario as I do not know how to explain the problem in any other ...
1
vote
1
answer
50
views
Chebyshev inequality comparison
Given $n$ distinct values $(x_i)_{i=1}^{n}$ with mean $\mu$ and standard deviation $s$, for all $i$, we have $|x_i−\mu| ≤ s \sqrt{n − 1}$.
How does this inequality compare with Chebyshev inequality as ...
1
vote
1
answer
734
views
Efficiently selecting a random subset of size $m$ from a set of size $n$
This is a cross post of my question here on math.se.
I have a list of $n$ items and would like to randomly select an $m$ set from it efficiently (in terms of time complexity). Also, I want all ...
2
votes
0
answers
53
views
In a machine learning system, why use differentially private SGD if our input data is already perturbed by a DP mechanism?
I'm trying to implement my own version of a deep neural network with differential privacy to preserve the privacy of the parties involved in the training dataset.
I'm using the method by Abadi et al. ...
0
votes
1
answer
46
views
Gaussian distribution with condition?
What does this expression mean?
Normal distribution with condition
I am reading a research paper and found the following expression (Eq.28 in the paper below).
It means a Gaussian distribution, but ...
1
vote
2
answers
68
views
Algorithm for Estimating Number of Unique Monthly Visitors
Is there a way to estimate the number of unique monthly visitors to a site based on a limited sample of one week of data? I have information about when a given user visited the site. This isn't as ...
9
votes
1
answer
2k
views
What is the optimal algorithm for playing the hangman word game?
Suppose we are playing the game hangman. My opponent and I both have access to the dictionary during the game. My opponent picks a word from the dictionary with knowledge of the algorithm which I will ...
1
vote
1
answer
40
views
Minimizing Trimmed Distances
I'm trying to understand what is the minimum of the following function,
$$
f(\mu) = \frac{1}{n}\sum_{i=1}^n F\left(\frac{\pi(i)}{n}\right) (x_i - \mu )^2
$$
where $F$ is a step function that assign $...
0
votes
0
answers
33
views
Does this problem have a formal name?
I have come across the following problem but am unable to understand the solution for it. Hence I would like to know if it has a formal name then, I can search for it and read about it in more detail. ...
0
votes
0
answers
62
views
How to find clusters of a set of points in n-dimensional space that are furthest from unwanted points
I have a list of 25 points and their coordinates in a 512-dimensional space. I have 8 target points and 17 points I need to avoid (the 17 points to avoid also have differences in priority of how ...
1
vote
1
answer
317
views
Amount of expected loop iterations when searching an array by random index
Lets say we have an array A of size n. It has 1 as its first index and n as its last index. It contains a value x, with x occurring k times in A where 1<=k<=n
If we have a search algorithm like ...
0
votes
0
answers
17
views
Kullback-Liebler Divergence
For $P(x)=N(\mu,\sigma^2)$ and $Q(x)=N(0,1)$ I am supposed to calculate $KL(P(x)||Q(x))$, here is what I did
\begin{align*}
KL(P(x)||Q(x)) & = \int P(x) \cdot \log\left(\frac{P(x)}{Q(x)}\...
3
votes
1
answer
166
views
Measures of performance in machine learning
I'm new to machine learning and struggle to interpret the results I get from different measures of performance. If for several prediction models I have e.g. accuracy, precision, recall, F1, FPR, and ...
1
vote
0
answers
110
views
Suggestion for a good statistics book for computer scientists, in preparation of machine learning
We (a group of CS postdocs and Ph.D. students) are starting a shared reading of a machine learning book ("An Introduction to Statistical Learning", James, Witten, Hastie, Tibshirani).
Before diving ...
2
votes
0
answers
63
views
The No-Free-Lunch Theorem and K-NN consistency
In computational learning, The NFL theorem states that there is no universal learner. For every learning algorithm , there is a distribution that causes the learner to output a hypotesis with a large ...
3
votes
0
answers
31
views
What is the purpose of standardization in machine learning?
I'm just getting started with learning about K-nearest neighbor and am having a hard time understanding why standardization is required. Reading through, I came across a section saying
When ...
3
votes
1
answer
65
views
Estimation of the number of solutions by Counting
This is a question from a quantum computation textbook.
Consider a classical algorithm for counting the number of solutions to a problem. The algorithm samples uniformly and independently $k$ times ...
0
votes
0
answers
70
views
What kind of standard deviation must be used in optimization algorithms?
I would like to ask about the standard deviation of objective function value.
There are two types of standard deviations:
Population standard deviation
Sample standard deviation
In metaheuristic ...
0
votes
1
answer
78
views
Assumption of a generation of the dataset by a probability distribution
Consider the following paragraph from the deeplearningbook
The training and test data are generated by a probability distribution
over datasets called the data-generating process. We typically ...
0
votes
2
answers
127
views
How can we get small test error reducing only train error?
My question is about mathematical part of machine learning algorithms, especially about using it in neural networks. We train network reducing train error and I was thinking about how then test error ...