What is `lr_policy` in Caffe?

Question

I just try to find out how I can use Caffe. To do so, I just took a look at the different .prototxt files in the examples folder. There is one option I don't understand:

# The learning rate policy
lr_policy: "inv"

Possible values seem to be:

• "fixed"
• "inv"
• "step"
• "multistep"
• "stepearly"
• "poly"

Could somebody please explain those options?

Solution

It is a common practice to decrease the learning rate (lr) as the optimization/learning process progresses. However, it is not clear how exactly the learning rate should be decreased as a function of the iteration number.

If you use DIGITS as an interface to Caffe, you will be able to visually see how the different choices affect the learning rate.

fixed: the learning rate is kept fixed throughout the learning process.

---

inv: the learning rate is decaying as ~1/T

---

step: the learning rate is piecewise constant, dropping every X iterations

---

multistep: piecewise constant at arbitrary intervals

---

You can see exactly how the learning rate is computed in the function SGDSolver<Dtype>::GetLearningRate (solvers/sgd_solver.cpp line ~30).

---

Recently, I came across an interesting and unconventional approach to learning-rate tuning: Leslie N. Smith's work "No More Pesky Learning Rate Guessing Games". In his report, Leslie suggests to use lr_policy that alternates between decreasing and increasing the learning rate. His work also suggests how to implement this policy in Caffe.

Comments

[Armin Meisterhirn]

No More Pesky ... is apparently similar in performance to using adadelta. There are now multiple adaptive schemes available in Caffe.