For an unconstrained convex minimization problem $$\begin{align} \text{minimize} \quad f(x) \end{align}$$ it is a well known that for a sufficiently small fixed step size $\alpha$, the gradient descent procedure defined by $$\hat x \equiv x - \alpha \nabla f(x)$$ converges to a minimum.

Now consider a convex minimization problem with a simple sign constraint: $$\begin{align} \text{minimize} \quad &f(x) \\ \text{subject to} \quad &x \geq 0 \end{align}$$ If we apply gradient descent with fixed step size, then some $x_i$ may turn negative. A common way to prevent this is using clamping, i.e. taking $$\hat x_i \equiv \max\lbrace{0, x_i - \alpha \nabla_{x_i} f(x) \rbrace}$$ (Notice that it is distinct from projected gradient descent, which minimizes the Euclidean distance between $\hat x$ and $x - \alpha \nabla f(x)$.)

  • Is there a canonical name for this gradient descent with fixed step size and clamping?
  • What are the conditions for its convergence?

I took the name "clamping" from this SE question; it doesn't seem to be a widely used term.

In computational economics, there is an analogous gradient process called tatonnement wherein prices adjust in the direction of excess demand but "clamp" to zero if the excess demand is highly negative. This is known to converge if the equivalent of $\nabla f$ satisfies a property called gross substitutability and the step size is sufficiently small. A canonical reference is H. Uzawa, 1960, “Walras’ tâtonnement in the theory of exchange,” The Review of Economic Studies 27, no. 3: 182–94.


1 Answer 1


The canonical name is called "projected gradient descent". Does the max operator not seem like a projection operator to you? It is taking a value, looks at it, and if it is negative, it projects it to the non-negative real line. So the clamping operation is the projector operator for the non-negative real line (and its generalizations).

Convergence for projected gradient descent seems to be well studied. Although I am not sure if its so obvious for fixed step-size. They usually require some diminishing stepsize.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .