@@ -58,7 +58,7 @@ jobs:
         working-directory: ${{ runner.temp }}
         env:
           RISKLABAI_REPOSITORY_ROOT: ${{ github.workspace }}
-        run: python -B -c "import os; from pathlib import Path; import RiskLabAI; origin = Path(RiskLabAI.__file__).resolve(); repository = Path(os.environ['RISKLABAI_REPOSITORY_ROOT']).resolve(); assert not origin.is_relative_to(repository), (origin, repository); assert RiskLabAI.__version__ == '3.1.0'; print(origin)"
+        run: python -B -c "import os; from pathlib import Path; import RiskLabAI; origin = Path(RiskLabAI.__file__).resolve(); repository = Path(os.environ['RISKLABAI_REPOSITORY_ROOT']).resolve(); assert not origin.is_relative_to(repository), (origin, repository); assert RiskLabAI.__version__ == '3.2.0'; print(origin)"
       - name: Run the installed base package tests
         working-directory: ${{ runner.temp }}
         run: python -B -m pytest -q -p no:cacheprovider -c "$GITHUB_WORKSPACE/pyproject.toml" "$GITHUB_WORKSPACE/test"
@@ -91,7 +91,7 @@ jobs:
         working-directory: ${{ runner.temp }}
         env:
           RISKLABAI_REPOSITORY_ROOT: ${{ github.workspace }}
-        run: python -B -c "import os; from pathlib import Path; import RiskLabAI; origin = Path(RiskLabAI.__file__).resolve(); repository = Path(os.environ['RISKLABAI_REPOSITORY_ROOT']).resolve(); assert not origin.is_relative_to(repository), (origin, repository); assert RiskLabAI.__version__ == '3.1.0'; print(origin)"
+        run: python -B -c "import os; from pathlib import Path; import RiskLabAI; origin = Path(RiskLabAI.__file__).resolve(); repository = Path(os.environ['RISKLABAI_REPOSITORY_ROOT']).resolve(); assert not origin.is_relative_to(repository), (origin, repository); assert RiskLabAI.__version__ == '3.2.0'; print(origin)"
       - name: Run the installed optional-feature tests
         working-directory: ${{ runner.temp }}
         run: python -B -m pytest -q -p no:cacheprovider -c "$GITHUB_WORKSPACE/pyproject.toml" "$GITHUB_WORKSPACE/test"
 
@@ -3,7 +3,38 @@
 All notable changes to RiskLabAI.py are documented here.
 Format: [Keep a Changelog](https://keepachangelog.com/en/1.1.0/); versioning: [SemVer](https://semver.org/).
 
-## [3.1.0] - Unreleased
+## [3.2.0] - Unreleased
+
+### Added
+
+- Sequential bootstrap sampling, majorization-based optimization, financial
+  network clearing, and optimal execution utilities.
+- Ambiguity calculations for supplied priors, order-flow features, regime and
+  drift adaptation, and tail-risk estimation and backtesting.
+- Constrained and robust portfolio optimization and decision-focused losses.
+- Causal bounds, finite-stratum transport, design-based interference, and
+  inverse-probability and doubly robust policy evaluation.
+- PyTorch implementations of TimeGAN, deep hedging, and finite-grid neural SDEs.
+- Analytical examples, independent numerical checks, boundary validation, and
+  regression tests for the new methods.
+
+### Fixed
+
+- Sample-concurrency counting for events that begin before the selected window.
+
+### Scope and dependencies
+
+- The neural-SDE implementation uses an explicitly corrected fixed-coordinate
+  diffusion operator rather than the literal published nearest-frame
+  construction. Its guarantees are bounded to the implemented finite grid;
+  general dynamic no-arbitrage and empirical market performance are not claimed.
+- Required dependencies and supported Python versions are unchanged. PyTorch
+  is available through the existing `pde` extra. Selected new methods require
+  separately installed CVXPY, River, or PySensemakr; these remain optional.
+- The additions are Python implementations. Corresponding Julia additions and
+  book notebooks are deferred.
+
+## [3.1.0] - Historical preparation notes
 
 ### Added
 
 
@@ -2,7 +2,7 @@
 
 ## Supported environments
 
-RiskLabAI 3.1.0 supports:
+RiskLabAI 3.2.0 supports:
 
 - CPython 3.12, 3.13, and 3.14
 - NumPy `>=2.2,<3`
@@ -73,7 +73,7 @@ The groups provide:
 | Extra | Capability |
 |---|---|
 | `speed` | Numba acceleration |
-| `pde` | Deep-BSDE PDE solver using PyTorch |
+| `pde` | PyTorch support for Deep-BSDE, neural SDEs, TimeGAN, and deep hedging |
 | `synth` | synthetic-control utilities using QuantEcon |
 | `hpo` | hyperparameter tuning using Optuna |
 | `plot` | Matplotlib, Seaborn, and Plotly helpers |
@@ -100,6 +100,18 @@ Notes:
 - PyWavelets is part of the base installation because its absence changes
   analytical behavior rather than only disabling acceleration.
 
+## Additional method dependencies
+
+Selected methods added in 3.2.0 use CVXPY (constrained or robust optimization),
+River (ADWIN drift adaptation), or PySensemakr (sensitivity analysis).
+These packages are installed separately when needed; they are not base
+dependencies or named extras. Local Python 3.13 validation used CVXPY 1.9.3,
+River 0.26.1, and PySensemakr 0.0.8. This does not establish compatibility with
+every version of those packages or every supported interpreter.
+
+The 3.2.0 version is a release candidate until published. An unpinned PyPI
+installation retrieves the latest published version.
+
 ## Development installation
 
 Clone the repository, switch to a supported Python version, and run the