Questions tagged [time-complexity]

Computational complexity (aka time complexity) of an algorithm is the amount of time it needs to run as a function of the input size.

Filter by
Sorted by
Tagged with
0 votes
1 answer
21 views

Time complexity of Metropolis-Hastings and potential speed-up?

The MH algorithm essentially involves generating a sample destination state from a proposal distribution, computing the acceptance probability as a function of that sample, and checking whether a ...
1 vote
1 answer
15 views

A question on computational complexity of a numerical differentiation (equation (5.77)) in Bishop's Pattern Recognition and Machine Learning

In page 249 of Christopher M. Bishop's book "Pattern Recognition and Machine Learning", it is said Again, the implementation of such algorithms can be checked by using numerical ...
  • 101
0 votes
0 answers
26 views

Can anyone explain the time complexity of policy iteration algorithm?

I am trying to answer my assignment on the RL, and this is the first question in chapter 3. What is the time complexity of this algorithm? I have not found any clear explanation on the internet.
0 votes
1 answer
32 views

Computational Complexity of Marginalization

Suppose you have a joint probability distribution with M variables, each sampled from a set of cardinality N. Now, suppose you want to marginalize one of the variables. My guess right now about the ...
0 votes
0 answers
8 views

finding max set of components that are in the most groups

so let's say you have a dataset like: ...
3 votes
0 answers
26 views

Check if growth rate is worse than quadratic?

I was asked to move the question here (https://stackoverflow.com/questions/72692849/check-if-growth-rate-is-worse-than-quadratic) Let's say I have collected a dataset for estimating algorithmic ...
  • 131
0 votes
0 answers
64 views

Learning Algorithm Time & Sample Complexity

Let $X=R^{2}$. Let $u=\left(\frac{\sqrt{3}}{2},-\frac{1}{2}\right),\ w=\left(-\frac{\sqrt{3}}{2},-\frac{1}{2}\right),\ v=\left(0,1\right)$ and $C=H=\left\{h\left(r\right)=\left\{\left(x_{1},x_{2\ }\...
1 vote
1 answer
42 views

Stepwise model selection by AIC

I am learning about performing stepwise model selection by AIC and having some questions: What is the regularization parameter for step-AIC? In what way is forward step-AIC an evolution of univariate ...
  • 51
1 vote
0 answers
16 views

How to estimate the time and memory requirement for Sparse and variational Gaussian Process (SVGP) with minibatch approach

I'm implementing Sparse and variational Gaussian Process (SVGP) using GPflow library using minibatches. I've been trying to search for details regarding the time and memory complexities e.g. from this ...
  • 5,377
2 votes
0 answers
47 views

What are some well-known unbiased estimator of regression coefficient besides OLS estimator?

Is there any other unbiased estimator of regression coefficient than OLS? For instance, one might consider using unbiased estimator with less computational cost (since OLS involves matrix inversion)?
0 votes
0 answers
121 views

Computational Complexity of Gradient Boosting Decision Tree algorithm

Does anybody have an idea about the computational complexity of GBDT? I have only seen one paper report (Gradient Boosted Decision Trees for High Dimensional Sparse Output). It doesn't seem to be ...
0 votes
0 answers
79 views

Why does the slow version of TreeExplainer algorithm have a complexity of $TLM2^M$

I'm trying to understand the TreeExplainer algorithms better from From Local Explanations to Global Understanding with Explainable AI for Trees, but confused by how the time complexity is derived even ...
  • 1,037
0 votes
0 answers
28 views

The sum of $n$ zero-mean values is of $O(\sqrt{n})$ at max

Why is the sum of $n$ zero-mean values is of $O(\sqrt{n})$ at max? If $X$ is a random variable with a standard deviation $\sigma$, the standard deviation of $n$ such i.i.d. random variables is $\sqrt{...
  • 271
4 votes
1 answer
346 views

Why Are Neural Networks Considered "Expensive" to Train?

Recently, I was looking at the optimization functions required in training Kernel Based Methods compared to Neural Networks. 1) Kernel Methods: For instance, I was looking at the optimization in ...
  • 6,920
0 votes
0 answers
44 views

Handling densely observed functional data with R package fda in reasonable time

I am trying to use the fda package to analyze functional data that is densely observed; for example, for one function I have ~25,000 samples of that function. (To be more precise, I observe $f(t_i), 1 ...
  • 1,183
3 votes
2 answers
134 views

Time complexity of training SVM and C

I always though that a big (high) C will lead to a lower svm training complexity due to lower number of support vectors. I was very surprised to read a paper that states the opposite - aka, big Cs ...
  • 31
1 vote
0 answers
168 views

FLOPS calculation for deep architectures [duplicate]

Pleas guide me as I have exhausted all my efforts to do so. I need to have the FLOPS value for DeepID2+ which is an FR model and maybe a few more (DeepID, Deep Id2). No publication mentions it and I ...
  • 11
1 vote
0 answers
14 views

Why do faster (eg sparse) versions of Transformers focus on the query-key product?

A lot of recent research on Transformers has been devoted to reducing the cost of the self-attention mechanism: $ softmax(\frac{Q K^T}{\sqrt{d}}) V $, As I understand it, the runtime, assuming $\{Q, K,...
5 votes
2 answers
2k views

Do deep neural networks learn slower with the addition of more hidden layers?

The number of hidden layers increases the number of weights, also increases the terms in the back-propagation algorithm, i.e. more derivatives, hence more computation. Can we say that neural networks ...
4 votes
1 answer
259 views

When does complexity in machine learning algorithms actually become an issue?

I keep seeing complexities for machine learning algorithms e.g. Gaussian process implementation requires inversion of an nxn matrix so complexity is $O(n^3)$. When does complexity actually become an ...
2 votes
1 answer
113 views

Cost of Nested Cross-Validation

What is the cost of nested cross-validation in terms of the number of times the algorithm needs to perform a fit-evaluate step? Based on this description of the algorithm, I think the answer is: $n \...
1 vote
1 answer
1k views

What is Big-O complexity of classifying an image using CNN?

If i have an image consisting of n pixels what will be the complexity of classifying it using a convolutional neural network, expressed in big-o notation? (assuming my cnn is already trained)
0 votes
1 answer
2k views

KNN Complexity (Big O notation)

I need to show the Big O Notation for KNN algorithm. So I wanted to know the complexity of brute force KNN algorithm; and to make the graph do we have x-axis: input size, y-axis: the speed.
1 vote
0 answers
52 views

Computational advantage for soft-impute method over other methods

I am reading in the soft-imputing paper for low-rank-based matrix completion. They suggested another solution for $$\hat{Z} = \text{argmin}_Z\lVert X - Z \rVert_F^2 + \lambda \lVert Z \rVert_*$$ ...
  • 291
0 votes
1 answer
166 views

Run time complexity of nearest neighbor

A paper titled, "Efficient Neighbor Searching in Nonlinear Time Series Analysis (1996)" download link mentions that the time complexity for the naive NN approach is $N^2/2$ i.e., $O(N^2/2)$ ...
  • 221
1 vote
0 answers
128 views

A custom loss function which weights the loss depending on the age of data used

I was wondering if it is possible to weight the loss so that old data are evaluated less than new data. Lets say I have a product which has a trend in value spanning across decades, I want the model ...
1 vote
0 answers
438 views

what are the main differences between parametric and non-parametric machine learning algorithms?

I am interested in parametric and non-parametric machine learning algorithms, their advantages and disadvantages and also their main differences regarding computational complexities. In particular I ...
1 vote
0 answers
67 views

Fast marginalizations of a large Probability Mass Distribution

I will explain my problem through a simple example. Imagine that I am given a probability mass function over a system of 10 3-states discrete random variables ($V_1,V_2,...,V_{10}$). So, I have $3^{10}...
2 votes
1 answer
293 views

How to find the time complexity of an MLE based algorithm

How to calculate or what is the time complexity (big-Oh) for this method? Based on my understanding, MLE depends on number of datapoints, $N$ so time complexity for MLE is O(N). However, there are ...
  • 221
3 votes
1 answer
539 views

Why does forward selection only take $O(p^2)$ calls to the learning algorithm?

In http://cs229.stanford.edu/notes/cs229-notes-all/cs229-notes5.pdf pg 5, it states that forward search takes $O(p^2)$ (note the notes uses $n$ instead of $p$ for the number if independent variables). ...
  • 563
0 votes
0 answers
87 views

Computational Complexity of SPADE, GSP and others

Does anybody know the computational complexity of SPADE, GSP, FreeSpan and PrefixSpan algorithm. I would like to have a comparisson in between these algorithms.
0 votes
0 answers
36 views

A question on time complexity

Suppose I have a sequence of i.i.d. Bernoulli random variable $X_i$ with mean $0$ and variance $1$. For each $X_i$, it is $1$ w.p $1/2$ and $-1$ w.p $1/2$. My goal is to show $X_1+X_2+\cdots+X_n \...
1 vote
0 answers
54 views

sample complexity vs training cost (time complexity)

I have a question about the complexities. Consider two regression problems A, B and assume that A has lower sample complexity than B. A and B share the same target function but the loses are slightly ...
  • 111
1 vote
0 answers
14 views

Reference Request: Ratio of Computational Complexity to Predictive Quality

I remember coming across a "performance metric" which was the ratio between the MSE and the computational time. However, I can't seem to recall the name of this metric or a reference. Does anyone ...
  • 544
0 votes
0 answers
800 views

Time complexity of bagging and random forest

Most sources and books state that time complexity of a single decision tree for n points and d dimensions (features) is $O(d * n^2 * log(n))$, with clever caching and one time sorting it’s $O(d * n * ...
  • 229
1 vote
2 answers
32 views

Issues with training on a sample of training set?

I am training an SVM on highly imbalanced data. I have rectified this issue and my ML pipeline works just fine. I have allocated 70% of my dataset for training, however this takes an infeasible amount ...
  • 11
1 vote
1 answer
3k views

Why does R take so much time to run auto.arima(). How can I shorten the calculation time? [duplicate]

I have been trying to run analysis and model a ts series of Natural Gas spot prices. With data provided by the Qandl API. The whole analysis was working fine, however, I experiences issues with the ...
  • 13
0 votes
0 answers
65 views

complexity of empirical estimator

Assume we have i.i.d. data $x_{1}, \dots, x_{n}$ from discrete distribution. Then, let's us consider empirical estimator: $$ \hat{p}_{i} = \frac{ \sum_{j=1}^{n}1(x_{j}=i)}{n} $$ What is the ...
  • 386
6 votes
1 answer
5k views

What is the computational complexity of a 1D convolutional layer?

What is the complexity of a 1D convolutional layer?. I'm getting $\mathcal{O}(n \cdot k \cdot d)$, but in Attention Is All You Need, Vaswani et al. report that it is $\mathcal{O}(k \cdot n \cdot d^2 )$...
  • 5,982
2 votes
1 answer
158 views

Complexity associated with decision trees

According to the sklearn documentation on decision trees: The cost of using the tree (i.e., predicting data) is logarithmic in the number of data points used to train the tree. Could somebody ...
  • 21
2 votes
2 answers
322 views

Computational complexity in practice: predict execution time for a dataset

Suppose I know that the computational complexity of an algorithm is $\mathcal{O}(f(n))$ where $n$ is the sample size. Suppose I have two data sets with sizes $n_1$ and $n_2$. The data sets have no ...
3 votes
0 answers
49 views

(Teaching) references for computational complexity

Background: I am going to teach computational complexity (time complexity) within an introductory course in machine learning. I would like to gently introduce the notion of computational complexity ...
1 vote
1 answer
38 views

Neural Networks and a catalogue of the number of floating point operations in various types of statistical and time series models

I recently came across this paper Green AI. In the paper they discussed using floating point operation (FPO) count during testing and implementation of Neural Networks (NNs) to compare the ...
  • 1,429
1 vote
0 answers
85 views

How to compute largest values of random variables? [closed]

Suppose we have two discrete random variables and we want perform maximum operation to obtain the max PDF. We know that max of two independent random variables is: if Z = max(X,Y) ...
  • 11
1 vote
0 answers
1k views

Time complexity of batch gradient descent

I am read http://papers.nips.cc/paper/4937-accelerating-stochastic-gradient-descent-using-predictive-variance-reduction.pdf paper. It states that "Due to the poor condition number, the standard batch ...
  • 175
0 votes
1 answer
310 views

knn asymptotic complexity vs svm [duplicate]

I'm doing a little report about the KNN complexity vs SVM.. I would like to know your opinions.. I built this text according to my perspective searching in papers, websites, ppts etc: The reason ...
1 vote
1 answer
728 views

Who is more complex computationally knn or SVM? [closed]

I have trained two models using sklearn library in python.. My dataset was about 750 features, 250 features per class (three classes), I trained only one feature dimension (1-D array). This are the ...
8 votes
4 answers
19k views

What is the computational cost of gradient descent vs linear regression?

I know the computational costs for the closed form of linear regression is $O(n^3)$, but I can't find a similar cost comparison to gradient descent. There are some similar questions here with people "...
1 vote
1 answer
3k views

What does it mean for K mean problem to be NP hard and why?

Given a decision problem (a problem with yes or no answer), the problem is said to be NP-hard if there is an NP-complete problem Y, such that Y is reducible to X in polynomial time. Recall that NP-...
0 votes
1 answer
881 views

How to inform the space and time complexity of K-means, SOM and Hierachical clustering

In the paper I am writing, one of the reviewers asked for an "a simple computational complexity analysis or time computational demands of their method" My question is : Can I simply report the ...