Consider the same house rent prediction problem where you are supposed to predict priceof a house based on just its area. Suppose you have n samples with their respective areas,x(1), x(2), ... , x(n), their true house rents y(1), y(2),..., y(n). Let's say, you train a linear regres-sor that predicts f(x()) = 00 + 01x(e). The parameters 6o and 0, are scalars and are learnedby minimizing mean-squared-error loss with L2-regularization through gradient descent witha learning rate a and the regularization strength constant A. Answer the following questions.1. Express the loss function(L) in terms of x), y@), n, 0, 01, A.2. Compute L3. Compute4. Write update rules for 6, and O1

The solution to the above question is:

Consider the same house rent prediction problem where you are supposed to predict price of a house based on just its area. Suppose you have n samples with their respective areas, 2), ., x(n), their true house rents y(1), y(2),., y(n). Let's say, you train a linear regres- sor that predicts f(x) = 0 + 01x). The parameters 0, and 61 are scalars and are learned by minimizing mean-squared-error loss with L2-regularization through gradient descent with a learning rate a and the regularization strength constant A. Answer the following questions. ..... 1. Express the loss function(L) in terms of x), y(),n,0, 01, A. 2. Compute L 3. Compute ƏL 4. Write update rules for 0, and 01

Consider the same house rent prediction problem where you are supposed to predict price of a house based on just its area. Suppose you have n samples with their respective areas, 2), ., x(n), their true house rents y(1), y(2),., y(n). Let's say, you train a linear regres- sor that predicts f(x) = 0 + 01x). The parameters 0, and 61 are scalars and are learned by minimizing mean-squared-error loss with L2-regularization through gradient descent with a learning rate a and the regularization strength constant A. Answer the following questions. ..... 1. Express the loss function(L) in terms of x), y(),n,0, 01, A. 2. Compute L 3. Compute ƏL 4. Write update rules for 0, and 01

Operations Research : Applications and Algorithms

4th Edition

ISBN:9780534380588

Author:Wayne L. Winston

Publisher:Wayne L. Winston

Chapter11: Nonlinear Programming

Section11.1: Review Of Differential Calculus

Problem 7P

See similar textbooks

Similar questions

Consider a logistic regression classifier that implements the 2-input OR gate. At iteration t, the parameters are given by w0=0, w1=0, w2=0. Given binary input (x1,x2), output of logistic regression is given by 1/(1+exp(-w0-w1*x1-w2*x2)). What will be value of the loss function at t? What will be the values of w0, w1 and w2 at (t+1) with learning rate ɳ=1?
(control variates) Reproduce the class example of estimating int 0 ^ 1 2 dz 1+x by the MC approach using 100 uniform random variables and after that by using a control variate with function g(U) = 1 + U as suggested in class. Compare the results.
In R, write a function that produces plots of statistical power versus sample size for simple linear regression. The function should be of the form LinRegPower(N,B,A,sd,nrep), where N is a vector/list of sample sizes, B is the true slope, A is the true intercept, sd is the true standard deviation of the residuals, and nrep is the number of simulation replicates. The function should conduct simulations and then produce a plot of statistical power versus the sample sizes in N for the hypothesis test of whether the slope is different than zero. B and A can be vectors/lists of equal length. In this case, the plot should have separate lines for each pair of A and B values (A[1] with B[1], A[2] with B[2], etc). The function should produce an informative error message if A and B are not the same length. It should also give an informative error message if N only has a single value. Demonstrate your function with some sample plots. Find some cases where power varies from close to zero to near…
Solve in R programming language: Let the random variable X be defined on the support set (1,2) with pdf fX(x) = (4/15)x3. (a) Find p(X<1.25). (b) Find EX. (c) Find the variance of X.
Give an example of a random variable X : {b, c, d, e} → N (Natural Number) with expectation 2, where each of {b, c, d, e} has equal probability
Computer Science Implement a function in python that takes in parameters X (a set of data points) and k (number of neighbors to use / smoothing amount) to execute kernel density estimation using k-nn regression, and plot the resultant estimation and real densities. You may use any libraries you want (eg. numpy, matplotlib, sklearn, etc)
Use R to answer the following question According to the central limit theorem, the sum of n independent identically distributed random variables will start to resemble a normal distribution as n grows large. The mean of the resulting distribution will be n times the mean of the summands, and the variance n times the variance of the summands. Demonstrate this property using Monte Carlo simulation. Over 10,000 trials, take the sum of 100 uniform random variables (with min=0 and max=1). Note: the variance of the uniform distribution with min 0 and max 1 is 1/12. Include: 1. A histogram of the results of the MC simulation 2. A density plot of a normal distribution with the appropriate mean and standard deviation 3. The mean and standard deviation of the MC simulation. ps(plz do not use chatgpt)
We are intrested in predicting the percentage of people commuting to work by walking given some input variables. Each observation corresponds to a different city and each input variable summarizes some characteristic of a given city, such as density, urban sprawl and average income per capita. This is 1. not a machine learning problem. Only social scientists would be interested in such a problem. 2. both a classification and a regression problem as it depends on the way one codes the output variable as either 0, 1 or a a particular number in the [0,1] interval. 3. a regression problem. The output variable is continuous. 4. a classification problem. Walking to work is a discrete variable and can only take two values: to walk to work and not to walk to wor
Consider a real random variable X with zero mean and variance σ2X . Suppose that we cannot directly observe X, but instead we can observe Yt := X + Wt, t ∈ [0, T ], where T > 0 and {Wt : t ∈ R} is a WSS process with zero mean and correlation function RW , uncorrelated with X.Further suppose that we use the following linear estimator to estimate X based on {Yt : t ∈ [0, T ]}:ˆXT =Z T0h(T − θ)Yθ dθ,i.e., we pass the process {Yt} through a causal LTI filter with impulse response h and sample theoutput at time T . We wish to design h to minimize the mean-squared error of the estimate.a. Use the orthogonality principle to write down a necessary and sufficient condition for theoptimal h. (The condition involves h, T , X, {Yt : t ∈ [0, T ]}, ˆXT , etc.)b. Use part a to derive a condition involving the optimal h that has the following form: for allτ ∈ [0, T ],a =Z T0h(θ)(b + c(τ − θ)) dθ,where a and b are constants and c is some function. (You must find a, b, and c in terms ofthe…
In R-Studio Problem 2. Find a large set of integers (with at least 10000 observations) representing a sample drawn from real life (i.e., do not generate the numbers using some random number generator). First, explain what the integers correspond to in real life. Then, investigate the distribution of the integers, and also the first and last digits of the integers in this sample (For example, if the data has [34 65], plot the distribution of these numbers and also distribution of [3,6] and [4 ,5] separately). Does any of these digits follow a uniform distribution? Is this expected? If one of them does not follow a uniform distribution, what distribution does it follow? Can you explain why? [You can analyze the image age dataset provided by IMDB Wiki, which contains the ages of actors and actresses whose photos are present in the IMDB database. The data contains 459868 age values in years, ranging from 1 to 99 years.]
From 1965 to 1974, in U.S. there were M = 17, 857, 857 male livebirths and F = 16, 974, 194 female livebirths. We model the number of male livebirth as a binomial distribution with parameters size = M+F and prob = p. The following code computes the maximum likelihood estimator for p. M <- 17857857 F <- 16974194 ll <- function(p){ dbinom(M, size=M+F, prob=p, log=TRUE) } ps <- seq(0.01, 0.99, by = 0.001) ll.ps <- ll(ps) plot(ps, ll.ps, type='l') phat <- ps[which.max(ll.ps)] abline(v = phat, col='blue') Question: What can we learn from the plot?
From 1965 to 1974, in U.S. there were M = 17, 857, 857 male livebirths and F = 16, 974, 194 female livebirths. We model the number of male livebirth as a binomial distribution with parameters size = M+F and prob = p. The following code computes the maximum likelihood estimator for p. M <- 17857857 F <- 16974194 ll <- function(p){ dbinom(M, size=M+F, prob=p, log=TRUE) } ps <- seq(0.01, 0.99, by = 0.001) ll.ps <- ll(ps) plot(ps, ll.ps, type='l') phat <- ps[which.max(ll.ps)] abline(v = phat, col='blue') Question: An estimator for p, denoted by pˆ, is obtained by ps[which.max(ll.ps)]. Is this the maximum likelihood estimator? Why (explain the code)?