Handle deterministic predictions in Expected Improvement

[AHMETHAKANBEZIR1]

Fixes #621

When predictive std is zero, expected improvement is the positive part of the deterministic improvement, max(mean-y_max-xi,0). Avoid dividing by zero and use this result on zero-variance entries. Positive-variance entries retain the same formula and floating operation order. ProbabilityOfImprovement and its tie convention are outside this change.

Add focused tests for mixed zero/positive variance, xi=0/nonzero, scalar calls, and a real sklearn GaussianProcessRegressor with a fixed constant kernel and alpha=0. The positive-variance control uses numerical integration with scipy.stats.norm.expect instead of restating the closed-form implementation.

Validation (Windows CPU / Python 3.12):

• New regression on untouched master af8b928: 3 failures. Real GP gives mean=[1,1], std=[0,0], old EI=[nan,nan].
• Full collected test suite: 178 passed, 24 warnings with NumPy 2.5.3 / SciPy 1.18.1 / sklearn 1.9.1; package statement coverage 98%, modified executable lines covered.
• Full acquisition module: 68 passed in both the above environment and NumPy 1.26.4 / SciPy 1.16.3 / sklearn 1.7.2.
• Positive-std random controls are bitwise equal to the original formula; scalar/vector broadcasting checked.
• Full configured Ruff 0.12.3 pre-commit lint and format checks and git diff check pass. Development extra resolves Ruff 0.16.10, which reports eight existing lint diagnostics and two existing Markdown-format differences on both untouched master and this branch; no new diagnostics.

Source installed editable with the development extra. Documentation builds, gallery notebook execution, GPU and other Python versions were not run. The current notebook-test path collects no example notebooks; the full collected test result should not be read as gallery validation.

AI assistance: Codex autonomously implemented the fix and ran validation; no independent human review has occurred. Codex is recorded as a commit coauthor.

Summary by CodeRabbit

• Bug Fixes
  • Expected-improvement calculations now work reliably when a prediction has zero uncertainty, avoiding division errors and returning a non-negative acquisition value.
  • Calculations for predictions with uncertainty continue to use the existing expected-improvement behavior. This improves reliability when evaluating deterministic predictions, including cases where predictive variance is zero.

[coderabbitai]

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Advanced

Run ID: 3a2540b7-2bf4-4745-abc5-17a01bcfbebf

📥 Commits

Reviewing files that changed from the base of the PR and between af8b928 and 6e8a460.

📒 Files selected for processing (2)

• bayes_opt/acquisition.py
• tests/test_acquisition.py

Included review availability: This review used your included allowance. Your plan provides up to 8 included reviews per hour; 7 remain after this review.

---

📝 Walkthrough

Walkthrough

ExpectedImprovement.base_acq now avoids division by zero when predictive standard deviation is zero. It returns the nonnegative improvement for deterministic predictions and retains the expected-improvement formula for nonzero standard deviation. New tests cover zero-variance inputs and a deterministic Gaussian process.

Changes

Expected improvement

Layer / File(s)	Summary
Zero-variance calculation and regression tests bayes_opt/acquisition.py, tests/test_acquisition.py	ExpectedImprovement.base_acq returns the nonnegative improvement when standard deviation is zero. Tests cover xi values of 0.0 and 0.25, strict NumPy error handling, and a deterministic Gaussian process.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~8 minutes

Change: Bug fix · Severity of issue fixed: Low

Merge Risk: ⚪ Minimal · up to 6e8a4

Expected Improvement now returns deterministic positive-part improvement at zero predictive variance while retaining its existing calculation for positive variance. No actionable merge risk is evident in the reviewed change.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 33.33% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 6 functions across 2 files.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely describes the main change: handling deterministic predictions in Expected Improvement.
Linked Issues check	✅ Passed	The PR addresses the coding requirements in [#621]. ExpectedImprovement.base_acq avoids division by zero and returns the deterministic positive-part improvement for zero standard deviation. Positive…
Out of Scope Changes check	✅ Passed	The reviewed changes are limited to ExpectedImprovement.base_acq and focused tests in tests/test_acquisition.py. These changes directly implement and verify [#621]. No unrelated production behavio…

• Fix all pre-merge checks with AI

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create a new PR

• Autopilot · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Autopilot is currently an internal CodeRabbit preview.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 98.36%. Comparing base (af8b928) to head (6e8a460).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #622   +/-   ##
=======================================
  Coverage   98.36%   98.36%           
=======================================
  Files          10       10           
  Lines        1220     1221    +1     
=======================================
+ Hits         1200     1201    +1     
  Misses         20       20           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.