Use optax losses as backend for jaxopt losses.

PiperOrigin-RevId: 616117244

Author: mtthss

jaxopt/_src/loss.py

@@ -17,11 +17,11 @@
 from typing import Callable
 
 import jax
-from jax.nn import softplus
 import jax.numpy as jnp
-from jax.scipy.special import logsumexp
 from jaxopt._src.projection import projection_simplex, projection_hypercube
 
+from optax import losses as optax_losses
+
 
 # Regression
 
@@ -39,10 +39,7 @@ def huber_loss(target: float, pred: float, delta: float = 1.0) -> float:
   References:
     https://en.wikipedia.org/wiki/Huber_loss
   """
-  abs_diff = jnp.abs(target - pred)
-  return jnp.where(abs_diff > delta,
-                   delta * (abs_diff - .5 * delta),
-                   0.5 * abs_diff ** 2)
+  return optax_losses.huber_loss(pred, target, delta)
 
 # Binary classification.
 
@@ -56,12 +53,8 @@ def binary_logistic_loss(label: int, logit: float) -> float:
   Returns:
     loss value
   """
-  # Softplus is the Fenchel conjugate of the Fermi-Dirac negentropy on [0, 1].
-  # softplus = proba * logit - xlogx(proba) - xlogx(1 - proba),
-  # where xlogx(proba) = proba * log(proba).
-  # Use -log sigmoid(logit) = softplus(-logit)
-  # and 1 - sigmoid(logit) = sigmoid(-logit).
-  return softplus(jnp.where(label, -logit, logit))
+  return optax_losses.sigmoid_binary_cross_entropy(
+      jnp.asarray(logit), jnp.asarray(label))
 
 
 def binary_sparsemax_loss(label: int, logit: float) -> float:
@@ -77,33 +70,7 @@ def binary_sparsemax_loss(label: int, logit: float) -> float:
     Learning with Fenchel-Young Losses. Mathieu Blondel, André F. T. Martins,
     Vlad Niculae. JMLR 2020. (Sec. 4.4)
   """
-  return sparse_plus(jnp.where(label, -logit, logit))
-
-
-def sparse_plus(x: float) -> float:
-  r"""Sparse plus function.
-
-  Computes the function:
-
-  .. math::
-
-    \mathrm{sparse\_plus}(x) = \begin{cases}
-      0, & x \leq -1\\
-      \frac{1}{4}(x+1)^2, & -1 < x < 1 \\
-      x, & 1 \leq x
-    \end{cases}
-
-  This is the twin function of the softplus activation ensuring a zero output
-  for inputs less than -1 and a linear output for inputs greater than 1,
-  while remaining smooth, convex, monotonic by an adequate definition between
-  -1 and 1.
-
-  Args:
-    x: input (float)
-  Returns:
-    sparse_plus(x) as defined above
-  """
-  return jnp.where(x <= -1.0, 0.0, jnp.where(x >= 1.0, x, (x + 1.0)**2/4))
+  return jax.nn.sparse_plus(jnp.where(label, -logit, logit))
 
 
 def sparse_sigmoid(x: float) -> float:
@@ -144,8 +111,7 @@ def binary_hinge_loss(label: int, score: float) -> float:
   References:
     https://en.wikipedia.org/wiki/Hinge_loss
   """
-  signed_label = 2.0 * label - 1.0
-  return jnp.maximum(0, 1 - score * signed_label)
+  return optax_losses.hinge_loss(score, 2.0 * label - 1.0)
 
 
 def binary_perceptron_loss(label: int, score: float) -> float:
@@ -160,8 +126,7 @@ def binary_perceptron_loss(label: int, score: float) -> float:
   References:
     https://en.wikipedia.org/wiki/Perceptron
   """
-  signed_label = 2.0 * label - 1.0
-  return jnp.maximum(0, - score * signed_label)
+  return optax_losses.perceptron_loss(score, 2.0 * label - 1.0)
 
 # Multiclass classification.
 
@@ -175,13 +140,8 @@ def multiclass_logistic_loss(label: int, logits: jnp.ndarray) -> float:
   Returns:
     loss value
   """
-  logits = jnp.asarray(logits)
-  # Logsumexp is the Fenchel conjugate of the Shannon negentropy on the simplex.
-  # logsumexp = jnp.dot(proba, logits) - jnp.dot(proba, jnp.log(proba))
-  # To avoid roundoff error, subtract target inside logsumexp.
-  # logsumexp(logits) - logits[y] = logsumexp(logits - logits[y])
-  logits = (logits - logits[label]).at[label].set(0.0)
-  return logsumexp(logits)
+  return optax_losses.softmax_cross_entropy_with_integer_labels(
+      jnp.asarray(logits), jnp.asarray(label))
 
 
 def multiclass_sparsemax_loss(label: int, scores: jnp.ndarray) -> float:
@@ -272,5 +232,6 @@ def make_fenchel_young_loss(max_fun: Callable[[jnp.ndarray], float]):
   """
 
   def fy_loss(y_true, scores, *args, **kwargs):
-    return max_fun(scores, *args, **kwargs) - jnp.vdot(y_true, scores)
+    return optax_losses.make_fenchel_young_loss(max_fun)(
+        scores.ravel(), y_true.ravel(), *args, **kwargs)
   return fy_loss

jaxopt/loss.py

@@ -12,8 +12,11 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import jax
+sparse_plus = jax.nn.sparse_plus
+
 from jaxopt._src.loss import binary_logistic_loss
-from jaxopt._src.loss import binary_sparsemax_loss, sparse_plus, sparse_sigmoid
+from jaxopt._src.loss import binary_sparsemax_loss, sparse_sigmoid
 from jaxopt._src.loss import huber_loss
 from jaxopt._src.loss import make_fenchel_young_loss
 from jaxopt._src.loss import multiclass_logistic_loss

requirements.txt

@@ -1,4 +1,5 @@
 jax>=0.2.18
 jaxlib>=0.1.69
 numpy>=1.18.4
+optax>=0.2.2
 scipy>=1.0.0