The tag has no usage guidance.

learn more… | top users | synonyms

1
vote
1answer
20 views

hyper parameter optimization grid search issues

I keep running into the same problem while doing a grid search to optimize the C and gamma parameters of an SVC. Every time i do the grid search, the best values seem to occur at around C = 100000 ...
12
votes
5answers
719 views

What's in a name: hyperparameters

So in a Normal distribution, we have two parameters: mean $\mu$ and variance $\sigma^2$. In the book Pattern Recognition and Machine Learning, there suddenly appears a hyperparameter $\lambda$ in the ...
1
vote
0answers
27 views

Tuning Parameters for Boosting/Bagging/Random Forest

I want to use tree-based classifiers for my classifiaction problem. I'm thinking about bagging, boosting (AdaBoost, LogitBoost, RUSBoost) and Random Forest but I'm unsure about the tuning parameters, ...
0
votes
1answer
26 views

Does memory ever really matter for mini-batch size selection?

I'm new to machine learning, and am confused about some aspects of stochastic gradient decent. I've read in several places that, when using vectorized code, the reason that mini-batching in ...
0
votes
0answers
12 views

How is it correct to optimize a binary classifier output threshold with ROC and LPOCV?

Hello everyone and thank you in advance for you help! I'm building a screening tool with a machine learning algorithm. The model provides a probabilistic prediction (i.e. logistic regression, ...
1
vote
0answers
27 views

Neural Network: How are number of parameters calculated?

I have a question about how the number of parameters (and limit for now to the number of weights because there can be other parameters as well) is calculated. Consider the screenshot below from the ...
0
votes
0answers
14 views

Hyperparameter optimization on large dataset

I have a huge dataset and want to carry out regression, such as gradient boosting. The problem is that the dataset is huge and hyperparameter optimization is computational expensive, especially I use ...
0
votes
0answers
31 views

Prior distributions for Alpha and beta parameters of beta-binomial model

Say that I'm constructing a beta-binomial model. The data comes from a binomial distribution, I assume that the parameter came from a beta distribution (my prior belief) - yet I've noticed that the ...
1
vote
0answers
22 views

How likely is a hyperparameter search on a subset dataset going to be accurate on the full dataset?

I'm running a hyper parameter search over many options in a neural network, holding all but one fixed in turn until I come to rest on a reasonably optimal set of hyperparameters. I wonder if anyone ...
1
vote
1answer
19 views

Why does the Hyperparameter optimization method GridSearch suffer from the curse of dimensionality?

An example accompanied by explanation is needed since I am a complete noob in these area.
0
votes
0answers
9 views

Heuristically, to what extent do estimator's hyperparameter tunning influence results?

Let's supose that you: Want to take a first insight into a predictive problem (classification, regression, clustering etc.). Don't need the optimal solution, just get an idea of how good the ...
0
votes
1answer
27 views

Can different classification methods be compared in the same manner as models during hyper-parameter tuning?

If I would like to choose between different classifiers, e.g. support vector machines (SVM) and boosted trees, based on their generalization performance, can I do this in the same way as I would do ...
0
votes
0answers
5 views

What does “Clamping factor” stands for in unsupervised label propogation?

see for example in sklearn package: http://scikit-learn.org/stable/modules/generated/sklearn.semi_supervised.LabelPropagation.html see reference http://pages.ucsd.edu/~ztu/publication/iccv13_dlp.pdf ...
1
vote
0answers
27 views

Why is there a perceived difference between a Bayesian Hyperparameter and a Machine Learning Hyperparameter?

There are two seperate Wikipedia articles for the term Hyperparameter. One for Bayesian statistics and another for machine learning methods. Why is this so? BOTH of these definitions imply that ...
1
vote
0answers
12 views

deriving shrinkage factors for beta binomial distributions

I am confused by the derivation of shrinkage factors currently online on sites like Wikipedia and ProbWiki. To be on point, I am confused how we get quickly from $\theta_i \sim Beta(\mu,M)$ to ...
0
votes
0answers
17 views

How high is “high” and how low is “low” in Latent Dirichlet Allocation Alpha and Eta hyperparameters? - LDA

In relation to this question and answer, the default value for the Python LDA for alpha is 0.1 and ...
0
votes
1answer
114 views

Denoising Autoencoder not training properly

I've implemented a denoising autoencoder using TensorFlow. The code is here, there is also a command line script to launch it. The code seems to work, the cross-validation error is decreasing every ...
1
vote
1answer
103 views

Random forest low score on testing data (scikit-learn)

I am trying to train my model using Scikit-learn's Random forest (Regression) and have tried to use GridSearch with Cross-validation (CV=5) to tune hyperparameters. I fixed ...
11
votes
1answer
236 views

Advantages of Particle Swarm Optimization over Bayesian Optimization for hyperparameter tuning?

There's substantial contemporary research on Bayesian Optimization (1) for tuning ML hyperparameters. The driving motivation here is that a minimal number of data points are required to make informed ...
2
votes
1answer
29 views

Levels of “hyperparameterization” in Hierarchical Modeling

Suppose we have observations $y$ that we wish to model as having being randomly sampled from a distribution with parameter $\theta$. General Bayesian approach assumes a prior distribution over ...
0
votes
0answers
40 views

How could a hyperparameter grid search be visualised?

Consider a hyperparameter grid search that looks at the training and testing scores of an estimator with respect to multiple parameters like training epochs, number of nodes in layer 1, number of ...
0
votes
0answers
80 views

normalization to zero mean and variance one logistic regression & random forrests

i was just thinking how does normalization to 0 mean and variance 1 (using an affine linear mapping) can impact the classification accuracy and the choice of hyperparameters when using: logistic ...
0
votes
0answers
27 views

Different Loss Functions used when fitting the model and tuning hyperparameters with cross validation

I usually do a two step process to fit a model: Given a hyper-parameter, fit the model using some criterion, such as MLE, etc. Do a k-fold cross validation and try a different hyper-parameter on ...
0
votes
0answers
51 views

Is this wikipedia article about KNN contradicting itself regarding “non-parametric”?

I understood that KNN (K-Nearest-Neighbors) was non-parametric, after reading the beginning of the wikipedia article here: ...
0
votes
1answer
94 views

How to determine the number of iterations for Latent Dirichlet Allocation

I am performing Latent Dirichlet Allocation for 240 test documents (trained model with 3361 documents). I am using 150 iterations and 120 burn in iterations. Is there a specific way to determine ...
2
votes
0answers
503 views

How to use XGboost.cv with hyperparameters optimization?

I want to optimize hyperparameters of XGboost using crossvalidation. However, it is not clear how to obtain the model from xgb.cv. For instance I call ...
4
votes
2answers
91 views

How exactly does one marginalize over parameters in an N-dimensional likelihood?

I see no equations for the following, so I'm not sure exactly what they are talking about: "For each model, we determine the best fit parameters from the peak of the N-dimensional likelihood surface. ...
1
vote
1answer
33 views

Can I use cross validation on a subset of the training set to select hyperparameters?

I am using R, and I had a dataset with 400000 rows and 800 columns, training a random forest model with only 100 trees on this dataset will take me about 1 and half hour on my laptop. So I went on and ...
2
votes
1answer
42 views

“Continuity” of SVM as a function of hyperparameters

Suppose we have some (large enough) labeled training set and use some exhaustive cross-validation technique (e.g. leave one out) for tuning hyperparameters of SVM (with some nonlinear kernel). Is it ...
2
votes
1answer
108 views

Can I do hyper-parameters optimization before model selection?

For every N model: Split in test and train subsets(Using the same seed for every N model) Randomized Search of parameters with 5 k-folds on train subset Select the best estimator obtained after the ...
2
votes
1answer
90 views

Hyperparameters Optimisation in Gaussian process for regression

I am trying to perform Gaussian Process for regression. I chose the SE Kernel : $K(x_i,x_j) = \exp(-\frac{||x_i-x_j||^2}{l}) + \sigma_n\delta_{i,j}$. I begin by maximize the log-likelihood with ...
0
votes
1answer
73 views

Which is better, MSE or classification rate, for minimizing the likelihood of overfitting?

During the process of selecting a classification model (with cross validation), if one uses MSE of the predicted posterior probabilities as a selection criterion, then a model ...
3
votes
2answers
193 views

Visualization figures for the grid search optimization steps?

I am using grid search in order to find the best values for the SVM parameters (namely C and ...
1
vote
1answer
211 views

How to optimize RBF parameters $C,\gamma$ with KSVM method?

I want to find the best choice of $C$ and $\gamma$ parameters for Radial Basis Function kernel. I am using kernlab instead of e1071 library. So how can i optimize RBF parameters $C$ and $\gamma$ with ...
2
votes
3answers
220 views

Practical hyperparameter optimization: Random vs. grid search

I'm currently going through Bengio's and Bergsta's Random Search for Hyper-Parameter Optimization [1] where the authors claim random search is more efficient than grid search in achieving ...
0
votes
0answers
97 views

marginal likelihood in linear bayesian regression (in weight-space)

I want to tune the hyperparameters namely the target deviance $\sigma_y$ and weight deviance $\sigma_w$ in bayesian linear regression. The posterior distribution in level-1 inference which is ...
1
vote
1answer
1k views

Hyperparameter Tuning - What is possible in terms of accuracy gain?

A question from a newbie: I played around with parameter tuning (grid, random search) in R (caret, xgboost) and my observation is as follows: in terms of accuracy gains I was able to get 3 - 7% but ...
2
votes
2answers
375 views

Nested cross-validation - how is it different from model selection via kfold CV on the training set?

I often see people talking about 5x2 cross-validation as a special case of nested cross validation. I assume the first number (here: 5) refers to the number of folds in the inner loop and the second ...
0
votes
0answers
30 views

set SVM parameter range values for tuning [duplicate]

I am newbie to using svm for classification. I want to tune svm parameters by .TrainAutofunction in EmguCV. But I don't know what are the range(min-max value) of below parameters that I should give to ...
0
votes
1answer
40 views

Marginal Likelihood of a Gaussian Process Model, Duplicate entries in kernel matrix

I am trying to fit a Gaussian process model using the toolbox and I got stuck in the following problem. Assuming that I have some duplicated data points in my training data, then those will map to ...
4
votes
1answer
163 views

Is an SVM's (maximum) likelihood uniquely defined as a function of hyperparameters?

I think that I must be reading this paragraph (below) incorrectly. Note that both types of evidence that we have defined in general depend on the inverse noise level $C$ and the kernel $K(x, ...
9
votes
4answers
898 views

How bad is hyperparameter tuning outside cross-validation?

I know that performing hyperparameter tuning outside of cross-validation can lead to biased-high estimates of external validity, because the dataset that you use to measure performance is the same one ...
1
vote
1answer
481 views

How to select hyperprior distribution for Beta distribution parameter?

I have a parameter $\theta$ whose value should lie between $(0,1)$. Therefore, I am assuming the prior distribution of $\theta$ to be a beta distribution with hyper-priors $\alpha$ and $\beta$ ie. ...
4
votes
1answer
297 views

Understanding the effect of hyperparameters in machine learning experiments

In machine learning every algorithm has a set of hyperparameters which needs to be optimized for best prediction performance. The simplest method for this optimization is called grid search which ...
1
vote
0answers
113 views

Can you take a DNN that was trained without regularization, and continue training it with regularization?

If I've trained a DNN with out any regularization methods (e.g. weight decay, dropout etc.) and reached a good training error, can I somehow take that learned net and fine tune it with regularization? ...
3
votes
1answer
133 views

The role of $\gamma$ & $C$ in SVM

I'm using support vector machine method with the Gaussian kernel. Is it true that $\gamma$ and $C$ are hyper parameters of SVM? What is their role exactly?
1
vote
2answers
87 views

Are classifier hyperparameters selected within cross-validation or not?

I was reading this question about selecting hyper-parameters for a support vector machine classifier, where grid-search is presented as one option. Which one is correct, either ...
0
votes
1answer
224 views

Estimating correlation hyperparameters of a Gaussian Process

I have an actual function that I need to simulate using a GP model. I've not done this before so I'm unclear of the steps. I have used the true function at different values of the inputs ($\vec X1, ...
2
votes
1answer
192 views

Dirichlet Process Hyperparameter Estimation with Sampling

I have a dirichlet process for which I need to learn the concentration (strength) hyperparameter (with gamma prior on it). One way of doing is via maximizing the Likelihood. Another way of doing this ...
2
votes
1answer
219 views

What is meant by effective parameters in machine learning

My question might be a bit ambiguous, but I started to wonder what does the "effective parameters" mean in machine learning? I have heard few professors of machine learning in my university talk about ...