3.2 Parameter estimation
Maximum likelihood estimation
For independent observations x1, …, xn, the log-likelihood is written as
ℒ(θ) = ∑ⁿᵢ₌₁ log p(xᵢ | θ)(7)
We obtain the estimator by maximizing Equation 7 with respect to the model parameters.
Ink2TeXiPhone app · Handwriting → LaTeX
See how it works.
Ink2TeX turns a photographed page of handwritten mathematics into an editable LaTeX document and a typeset PDF, on the phone. Walk through the app below, then tell us whether you would use it.
Walk through the appNo account, nothing to install, nothing uploaded.
The app, step by step
Scan a page. Get a document.
Pick a page
Theorem 2.1 (Spectral decomposition)
Let A ∈ ℝⁿˣⁿ be symmetric. Then
A = Q Λ Qᵀ
where Q is orthogonal and Λ is diagonal.
rank(A) = #{ i : λᵢ ≠ 0 }
Three questions
Three questions
Would this be useful to you?
We are building Ink2TeX for people who write mathematics by hand. Three taps tell us whether we are building the right thing — no account, no obligation.
How do you get handwritten mathematics into LaTeX today?
Would you use this on your phone?
What would it be worth to you?
Want it when it ships?One email when Ink2TeX reaches the App Store. Nothing else, ever.
One equation, two forms
From a fixed page to editable source
\begin{equation}
\mathcal{L}(\theta)
= \sum_{i=1}^{n} \log p(x_i \mid \theta)
\end{equation}Also on the web
Convert a research paper instead
A separate web prototype: give it an arXiv link or a text-based PDF and it returns the mathematics as LaTeX. The iPhone app above is the one that reads handwriting.