[MXNET-978] Higher Order Gradient Support arcsin, arccos.

src/operator/tensor/elemwise_unary_op_trig.cc

@@ -188,7 +188,31 @@ The storage type of ``arcsin`` output depends upon the input storage type:
 .set_attr<nnvm::FGradient>("FGradient", ElemwiseGradUseIn{ "_backward_arcsin" });
 
 MXNET_OPERATOR_REGISTER_BINARY_WITH_SPARSE_CPU_DR(_backward_arcsin,
- unary_bwd<mshadow_op::arcsin_grad>);
+ unary_bwd<mshadow_op::arcsin_grad>)
+.set_attr<nnvm::FGradient>("FGradient",
+ [](const nnvm::NodePtr& n, const std::vector<nnvm::NodeEntry>& ograds) {
+ // ograds[0]: head_grad_grads (dL/dxgrad)
+ // inputs[0]: dL/dy
+ // inputs[1]: x (ElemwiseGradUseIn)
+ // f(x) = arcsin(x)
+ // n: f'(x) = 1/(1-x^2)^1/2
+ // f''(x) = f'(x) * x/(1-x^2)
+ // Note: x/(1-x^2) = x * f'(x)^2
+ auto dydx = n->inputs[0];
+ auto x = n->inputs[1];
+ auto dydx_mul_grad_x = nnvm::NodeEntry{n};
+ auto op = mxnet::util::NodeOpGen{n};
+
+ auto x_grad = op.div(dydx_mul_grad_x, dydx);
+ auto x_grad_square = op.square(x_grad);
+ auto x_grad_square_mul_x = op.mul(x_grad_square, x);
+ auto x_grad_grad = op.mul(dydx_mul_grad_x, x_grad_square_mul_x);
+
+ std::vector<nnvm::NodeEntry> ret;
+ ret.emplace_back(op.mul(ograds[0], x_grad));
+ ret.emplace_back(op.mul(ograds[0], x_grad_grad));
+ return ret;
+ });
 
 // arccos
 MXNET_OPERATOR_REGISTER_UNARY_WITH_SPARSE_DR(arccos, cpu, mshadow_op::arccos)
@@ -207,7 +231,32 @@ The storage type of ``arccos`` output is always dense
 .set_attr<nnvm::FGradient>("FGradient", ElemwiseGradUseIn{ "_backward_arccos" });
 
 MXNET_OPERATOR_REGISTER_BINARY_WITH_SPARSE_CPU_DR(_backward_arccos,
- unary_bwd<mshadow_op::arccos_grad>);
+ unary_bwd<mshadow_op::arccos_grad>)
+.set_attr<nnvm::FGradient>("FGradient",
+ [](const nnvm::NodePtr& n, const std::vector<nnvm::NodeEntry>& ograds) {
+ // ograds[0]: head_grad_grads (dL/dxgrad)
+ // inputs[0]: dL/dy
+ // inputs[1]: x (ElemwiseGradUseIn)
+ // f(x) = arccos(x)
+ // n: f'(x) = -1/(1-x^2)^1/2
+ // f''(x) = f'(x) * x/(1-x^2)
+ // Note: x/(1-x^2) = x * f'(x)^2
+ auto dydx = n->inputs[0];
+ auto x = n->inputs[1];
+ auto dydx_mul_grad_x = nnvm::NodeEntry{n};
+ auto op = mxnet::util::NodeOpGen{n};
+
+ auto x_grad = op.div(dydx_mul_grad_x, dydx);
+ auto x_grad_square = op.square(x_grad);
+ auto x_grad_square_mul_x = op.mul(x_grad_square, x);
+ auto x_grad_grad = op.mul(dydx_mul_grad_x, x_grad_square_mul_x);
+
+ std::vector<nnvm::NodeEntry> ret;
+ ret.emplace_back(op.mul(ograds[0], x_grad));
+ ret.emplace_back(op.mul(ograds[0], x_grad_grad));
+ return ret;
+ });
+
 
 // arctan
 MXNET_OPERATOR_REGISTER_UNARY_WITH_RSP_CSR(arctan, cpu, mshadow_op::arctan)

tests/python/unittest/test_higher_order_grad.py

@@ -114,6 +114,44 @@ def grad_grad_op(x):
  array, tanh, grad_grad_op, rtol=1e-6, atol=1e-6)
 
 
+@with_seed()
+def test_arcsin():
+ def arcsin(x):
+ return nd.arcsin(x)
+
+ def grad_grad_op(x):
+ return x / nd.sqrt((1-x**2)**3)
+
+ for dim in range(1, 5):
+ shape = rand_shape_nd(dim)
+ array = random_arrays(shape)
+ # Hack: Decrease std_dev to make
+ # sure all elements
+ # are in range -1 to 1
+ # i.e. Domain of arcsin
+ array *= 0.2
+ check_second_order_unary(array, arcsin, grad_grad_op)
+
+
+@with_seed()
+def test_arccos():
+ def arccos(x):
+ return nd.arccos(x)
+
+ def grad_grad_op(x):
+ return -x / nd.sqrt((1-x**2)**3)
+
+ for dim in range(1, 5):
+ shape = rand_shape_nd(dim)
+ array = random_arrays(shape)
+ # Hack: Decrease std_dev to make
+ # sure all elements
+ # are in range -1 to 1
+ # i.e. Domain of arccos
+ array *= 0.2
+ check_second_order_unary(array, arccos, grad_grad_op)
+
+
 @with_seed()
 def test_arctan():
  def arctan(x):