differentiable_programming.rst

Differentiable programming

Please check out the DiffTaichi paper and video to learn more about Taichi differentiable programming.

The DiffTaichi repo contains 10 differentiable physical simulators built with Taichi differentiable programming.

Note

Unlike tools such as TensorFlow where immutable output buffers are generated, the imperative programming paradigm adopted in Taichi allows programmers to freely modify global tensors. To make automatic differentiation well-defined under this setting, we make the following assumption on Taichi programs for differentiable programming:

Global Data Access Rules:

• If a global tensor element is written more than once, then starting from the second write, the write must come in the form of an atomic add (“accumulation", using ti.atomic_add or simply +=).
• No read accesses happen to a global tensor element, until its accumulation is done.

Kernel Simplicity Rule: Kernel body consists of multiple simply nested for-loops. I.e., each for-loop can either contain exactly one (nested) for-loop (and no other statements), or a group of statements without loops.

Example:

@ti.kernel
def differentiable_task():
  for i in x:
    x[i] = y[i]

  for i in range(10):
    for j in range(20):
      for k in range(300):
        ... do whatever you want, as long as there is no loops

  # Not allowed. The outer for loop contains two for loops
  for i in range(10):
    for j in range(20):
      ...
    for j in range(20):
      ...

Taichi programs that violate this rule has an undefined gradient behavior.

Note that currently the Taichi compiler is not able to detect such violation automatically - please be careful.

A few examples with neural network controllers optimized using differentiable simulators and brute-force gradient descent:

Documentation WIP.