Skip to content

Start Learning Gravity

Scrollable formula. Use Left and Right arrow keys to reveal hidden content.

Scrollable table. Use Left and Right arrow keys to reveal additional columns.

Scrollable figure. Use Left and Right arrow keys to inspect hidden content.

Start with the work you want to be able to do. This page asks only the preparation questions that matter for that goal, then gives one of three results: ready, review, or repair. It does not ask about degrees, job titles, grades, or how long you have studied. You may always ignore its recommendation and choose any route directly.

This route requires no account or timer, uses no JavaScript, and stores no answers. If you want a private record, copy the short result table into your own notes.

If a copied or bookmarked result names a route that is no longer offered, that result is stale. Treat the named route as unavailable, assign no ready, review, or repair result, and choose again from the routes currently shown here. An unavailable route is neither displayed as a link nor replaced by a guessed recommendation.

Pick the row that best describes the next piece of work—not a permanent label for your background.

I want to…Begin hereReadiness check
form a first accurate picture of gravity and black holesFirst Encounter with Gravity and Black Holesnone; begin directly
prepare the mathematics and physics used in relativityMathematics and Physics for Relativity or the preparation directorynone; the path diagnoses each capability as it arises
learn graduate general relativityGraduate General Relativity Self-Study Path or the First Graduate Coursecalculus and linear algebra; differential equations; special relativity; tensor geometry
specialize in a branch of gravityLearning Pathsuse the selected path’s own prerequisite check
prepare for the Evidence and Open Problems seminarResearch Seminar: Evidence and Open Problems in Gravitation or the Research Atlasseminar common entry only: Einstein’s equation; elementary inference; selected technical cases add prerequisites
assemble a coherent courseCourses and Seminarscompare audience, outcomes, prerequisites, assignments, and capstone

The two goals with a readiness check are worked out below. For every prompt, try the task without notes and mark the result:

  • secure — you can complete it and explain why the method works;
  • uncertain — the main idea is familiar, but your explanation or execution has a gap; or
  • not yet — you cannot presently begin, or would have to imitate steps you could not justify.

Use the least favorable result honestly. A single not yet means repair. With no not yet but at least one uncertain, the result is review. All secure means ready for that goal’s stated entry tasks. Use this result to choose what to study next.

Graduate relativity uses four capabilities immediately and repeatedly. Work the four checks in order. Each one is deliberately small: it tests whether a tool is available, not whether you have already learned general relativity.

Let

A=(2112),v=(1−1).A=\begin{pmatrix}2&1\\1&2\end{pmatrix}, \qquad v=\begin{pmatrix}1\\-1\end{pmatrix}.

Find AvAv, identify the eigenvalue belonging to vv, and explain in one sentence why the eigenvalue is independent of multiplying vv by a nonzero constant. Then differentiate f(x,y)=x2eyf(x,y)=x^2e^y with respect to both variables.

Check the reasoning

Av=(1,−1)T=vAv=(1,-1)^T=v, so the eigenvalue is 11. If Av=λvAv=\lambda v, then A(cv)=cAv=λ(cv)A(cv)=cAv=\lambda(cv) for every nonzero cc; rescaling changes the representative vector, not the eigendirection or eigenvalue. The derivatives are ∂xf=2xey\partial_x f=2xe^y and ∂yf=x2ey\partial_y f=x^2e^y.

Mark secure only if both the calculation and rescaling explanation are clear. If the calculation is right but the explanation is shaky, mark uncertain.

For review or repair, use Calculus and Linear Algebra for Relativity.

For ω>0\omega\gt0, solve

y′′+ω2y=0,y(0)=a,y′(0)=0,y''+\omega^2y=0, \qquad y(0)=a, \qquad y'(0)=0,

and name a quantity built from yy and y′y' that remains constant.

Check the reasoning

The solution is y(x)=acos⁡(ωx)y(x)=a\cos(\omega x). One conserved quantity is E=12[(y′)2+ω2y2]E=\tfrac12[(y')^2+\omega^2y^2]; differentiating gives E′=y′(y′′+ω2y)=0E'=y'(y''+\omega^2y)=0. A correct formula without the substitution or conservation check is evidence for uncertain, not secure.

For review or repair, use Differential Equations and Qualitative Dynamics.

In units with c=1c=1, two events have coordinate separation Δt=5\Delta t=5 and Δx=3\Delta x=3 in an inertial frame. Classify the separation, compute the proper-time interval using the site’s (−,+,+,+)(-,+,+,+) convention, and say which part of your answer every inertial observer must agree on.

Check the reasoning

Δs2=−Δt2+Δx2=−16\Delta s^2=-\Delta t^2+\Delta x^2=-16, so the separation is timelike and Δτ=−Δs2=4\Delta\tau=\sqrt{-\Delta s^2}=4. Other inertial observers assign different coordinate differences, but agree on the interval, its timelike character, and the proper time between the events. If signs were guessed rather than connected to the declared convention, mark uncertain.

For review or repair, use Mechanics and Special Relativity Diagnostic.

Explain why the components of a vector change under a coordinate transformation while the geometric vector does not. Then say why Christoffel symbols can vanish at a point in a suitable coordinate system even when the Riemann curvature there is nonzero.

Check the reasoning

A vector is a coordinate-independent tangent-space element; its components are coefficients in a chosen basis, and the basis changes with coordinates. Christoffel symbols are connection coefficients rather than tensor components, so their inhomogeneous transformation law permits them to vanish at one point. Curvature is tensorial and depends on derivatives and quadratic combinations of the connection, so nonzero curvature cannot be transformed away. Mark secure only if both distinctions are explicit.

For review or repair, use Tensor Geometry Bridge.

ResultNext action
readyBegin the Graduate General Relativity Self-Study Path. Choose the First Graduate Course instead when you want fixed assignments and assessment criteria.
reviewOpen the linked module for every uncertain capability, attempt its diagnostic and transfer problem, then return to the graduate path.
repairBegin with the first linked module marked not yet. Complete its review sequence and changed-setting transfer problem before checking the next gap. Then return to the graduate path.

Multiple gaps do not require completing every preparation page. Review the relevant topics in the order above because later checks use earlier tools. If a module’s opening diagnostic is already secure, take its transfer problem and move on.

Readiness for the Evidence and Open Problems seminar

Section titled “Readiness for the Evidence and Open Problems seminar”

This is the common entry diagnostic for one seminar, not a general test of research readiness. It checks whether you can recover one classical-gravity argument and interpret elementary Bayesian evidence language used across the seminar’s cases. Passing it does not establish readiness for research across gravity, nor does it supply the technical preparation for every case.

Before a bounded technical reproduction, the seminar applies the selected case’s own prerequisite profile. Advanced learning paths likewise own their branch-specific entry checks; readiness for one branch must not be inferred from the two common tasks below.

Write Einstein’s equation with a cosmological constant. Identify its geometric and matter sides, and explain why covariant conservation of stress-energy is consistent with the equation.

Check the reasoning

With the site’s convention,

Gμν+Λgμν=8πGTμν.G_{\mu\nu}+\Lambda g_{\mu\nu}=8\pi G T_{\mu\nu}.

The left side is geometric and the right side represents matter and non-gravitational fields. The contracted Bianchi identity gives ∇μGμν=0\nabla^\mu G_{\mu\nu}=0, while metric compatibility gives ∇μgμν=0\nabla^\mu g_{\mu\nu}=0; for constant Λ\Lambda, the equation therefore implies ∇μTμν=0\nabla^\mu T_{\mu\nu}=0. Mark uncertain if you can state the equation but cannot supply that chain of reasoning.

For review or repair, study Einstein Field Equations and check each convention before carrying a formula into a seminar case.

Suppose a model has parameter θ\theta and data dd. State the roles of the likelihood, prior, posterior, and evidence in Bayes’ theorem. Then explain in one sentence why an unmodeled selection effect can bias an inferred population.

Check the reasoning

Bayes’ theorem reads

p(θ∣d)=p(d∣θ)p(θ)p(d).p(\theta\mid d)=\frac{p(d\mid\theta)p(\theta)}{p(d)}.

p(d∣θ)p(d\mid\theta) is the likelihood, p(θ)p(\theta) the prior, p(θ∣d)p(\theta\mid d) the posterior, and p(d)p(d) the marginal likelihood or evidence. A selection process changes which systems enter the observed sample; if that detection probability is omitted, the observed distribution is mistaken for the underlying population. Mark secure only if the symbols and the selection argument are both interpretable in words.

For review or repair, use Probability and Inference for Gravity.

All secure means that the seminar’s two common entry tasks are available; it leads to Research Seminar: Evidence and Open Problems in Gravitation, not to a claim of domain-general research readiness. Any uncertain result calls for the linked review before the seminar; any not-yet result calls for repair. The selected case still applies its own prerequisite check before technical reproduction. If your goal is to inspect dated evidence without undertaking the seminar assignments, browse the Research Atlas directly.

A diagnostic can be wrong about what you need. You may know a tool in a notation unlike the prompt, have an accessibility reason to demonstrate it in another form, or be deliberately exploring ahead of your preparation. Choose the direct route whenever that serves the work, and use its first checkpoint as better evidence.

If you stop, retain only three facts: the goal you chose, the least secure capability, and the linked next page. On returning, repeat that capability’s small check. The route is designed to reconstruct its state from your work, without a browser history or saved profile.