Objective, constraint and regularity
The elementary single-constraint theorem assumes continuously differentiable functions nearby.[1]
For a regular constraint, directions orthogonal to∇g form the local tangent space. At an extremum, f has zero first-order change along every such feasible direction. Therefore∇f lies along the normal∇g, giving the multiplier condition. If∇g vanishes, this argument fails and the point needs separate inspection.[2]
Check all candidates
Solve the gradient equations together with g=c. Reject infeasible points and compare objective values or apply an appropriate local constrained test. A continuous objective on a nonempty compact feasible set attains global extrema. An unbounded set lacks that guarantee; it can have a minimum but no maximum.[1]
This convention makes∇xL=0 equivalent to∇f=λ∇g. A plus sign in L reverses the multiplier sign, not the feasible extrema.[2]
What the method does not claim
λ has objective-units divided by constraint-units, so its interpretation depends on scaling and Lagrangian sign. Several equality constraints need the appropriate independence of their gradients. Inequalities require additional optimality conditions; the equation here is neither a universal optimizer nor a full KKT method.[1]
Two worked examples
Quadratic cost on a line
- Given
- f=x2+y2, g=x+y=6.
- Find
- Find the global minimum and verify it.
Set up gradients and the constraint.
This gives x=y=3. An independent substitution classifies it.
Result: Minimum 18 at(3,3),λ=6; there is no maximum.
Check: The square is nonnegative and becomes arbitrarily large along the unbounded line.
Four candidates on a circle
- Given
- f=xy, g=x2+y2=8.
- Find
- Find maximum and minimum.
The constraint is regular because this circle excludes the origin.
Thus y=±x and x=±2.
Result: Maximum 4 for matching signs; minimum−4 for opposite signs.
Check: The inequalities(x−y)2≥0 and(x+y)2≥0 independently give−4≤xy≤4 on the circle.
Recall the method later with the formula’s flashcards.
10 flashcards · English materials
Already have an account? Sign in for flashcardsAvailable from Essential. A ready-made formula set, no AI credits required. Article exercises are not all included.
Five exercises with solutions
Calculate · Foundation
Minimize x2+y2 subject to x+y=10.
Show hint
The gradient equations give x=y.
Learning goal: Compute a regular constrained minimum.
Show solution and calculation
x=y=5 and f=50.
Substitution gives 2(x−5)2+50, confirming the global minimum.
Interpret · Application
Why does∇f parallel∇g indicate tangent level curves?
Show hint
Gradients are normal to regular level curves.
Learning goal: Explain geometry without claiming sufficiency.
Show solution and calculation
Parallel normals give the same tangent direction, explaining the necessary first-order condition without classifying the candidate.
Error analysis · Application
What is wrong with “Every multiplier solution is a minimum”?
Show hint
Use the circle example for xy.
Learning goal: Separate necessity from classification.
Show solution and calculation
The condition is necessary, not sufficient.
On x2+y2=8, four candidates give both maxima 4 and minima−4, so value comparison remains necessary.
Transfer · Challenge
What does f=x subject to x2+y2=0 show?
Show hint
The feasible set contains one point.
Learning goal: Test why regularity is required.
Show solution and calculation
Only(0,0) is feasible, so f is both globally minimal and maximal there.∇g=0 while∇f=(1,0), so noλ solves the multiplier equation.
The regular theorem is not contradicted.
Rearrange · Foundation
Does using L=f+μ(g−c) change the extrema?
Show hint
Compare with L=f−λ(g−c).
Learning goal: Compare sign conventions consistently.
Show solution and calculation
No.
Withμ=−λ, the equations agree.
Points and objective values stay the same; only the multiplier sign changes.
Check your understanding with the formula’s reference test.
7 test questions · English materials
Already have an account? Sign in for the testAvailable from Essential. A ready-made formula set, no AI credits required. Article exercises are not all included.
Useful next steps
Compute the normal vectors using Partial derivatives and gradient.
Zero first-order change along a feasible tangent is expressed by the Directional Derivative.
Inequalities in a linear model instead belong to Linear programming and simplex.
Sources for this explanation
Quanta · Updated November 8, 2026
- OpenStax Calculus Volume 3: 4.8 Lagrange Multipliers
- University of Stuttgart, Higher Mathematics (Stroppel): 4.6 Constrained extrema (translated source label)
Editorial responsibility: Quanta Study editorial team
Review scope
This explanation, 2 worked examples and 5 exercises are available in German and English. Structure, source references and mathematical notation undergo automated checks. Calculations, model assumptions and both language versions receive a separate AI review.
An independent, named human subject review has not yet been documented. Our exercise numbers are worked examples, rather than measurements published by the cited sources.
Editorial responsibility, sources and methodsChange history
- Added regular multiplier conditions, candidate checks, a singular counterexample and sign conventions in DE/EN.
Cite this reference page
Quanta. Lagrange multipliers. Updated November 8, 2026. https://quanta-study.de/en/formula-sheet/lagrange-multiplikatoren
When citing a specific scientific statement, also cite the specialist source linked to that statement.