Add Federated Reconstruction for Matrix Factorization tutorial.

PiperOrigin-RevId: 392999200

docs/_book.yaml

@@ -22,6 +22,8 @@ upper_tabs:
         path: /federated/tutorials/federated_learning_for_text_generation
       - title: "Tuning Recommended Aggregations for Learning"
         path: /federated/tutorials/tuning_recommended_aggregators
+      - title: "Federated Reconstruction for Matrix Factorization"
+        path: /federated/tutorials/federated_reconstruction_for_matrix_factorization
 
       - heading: Custom computations
       - title: "Building Your Own Federated Learning Algorithm"

docs/tutorials/BUILD

@@ -11,6 +11,7 @@ exports_files([
     "federated_learning_for_image_classification.ipynb",
     "federated_learning_for_text_generation.ipynb",
     "federated_learning_with_differential_privacy.ipynb",
+    "federated_reconstruction_for_matrix_factorization.ipynb",
     "federated_select.ipynb",
     "random_noise_generation.ipynb",
     "simulations.ipynb",

docs/tutorials/tutorials_overview.md

@@ -13,11 +13,15 @@ documentation can be found in the [TFF guides](../get_started.md).
 *   [Federated Learning for text generation](federated_learning_for_text_generation.ipynb)
     further demonstrates how to use TFF's FL API to refine a serialized
     pre-trained model for a language modeling task.
-
 *   [Tuning recommended aggregations for learning](tuning_recommended_aggregators.ipynb)
     shows how the basic FL computations in `tff.learning` can be combined with
     specialized aggregation routines offering robustness, differential privacy,
     compression, and more.
+*   [Federated Reconstruction for Matrix Factorization](federated_reconstruction_for_matrix_factorization.ipynb)
+    introduces partially local federated learning, where some client parameters
+    are never aggregated on the server. The tutorial demonstrates how to use the
+    Federated Learning API to train a partially local matrix factorization
+    model.
 
 **Writing custom federated computations**