Ad coefficient update for last seen ad

[ptjames]

This PR updates the default value of last_seen_ad_predictor_weight to 0. Recent AB testing demonstrates this updates leads to a significant increase in CTR.

Resolves brave/brave-browser#39326

Submitter Checklist:

• I confirm that no security/privacy review is needed and no other type of reviews are needed, or that I have requested them
• There is a ticket for my issue
• Used Github auto-closing keywords in the PR description above
• Wrote a good PR/commit description
• Squashed any review feedback or "fixup" commits before merge, so that history is a record of what happened in the repo, not your PR
• Added appropriate labels (QA/Yes or QA/No; release-notes/include or release-notes/exclude; OS/...) to the associated issue
• Checked the PR locally:
  • npm run test -- brave_browser_tests, npm run test -- brave_unit_tests wiki
  • npm run presubmit wiki, npm run gn_check, npm run tslint
• Ran git rebase master (if needed)

Reviewer Checklist:

• A security review is not needed, or a link to one is included in the PR description
• New files have MPL-2.0 license header
• Adequate test coverage exists to prevent regressions
• Major classes, functions and non-trivial code blocks are well-commented
• Changes in component dependencies are properly reflected in gn
• Code follows the style guide
• Test plan is specified in PR before merging

After-merge Checklist:

• The associated issue milestone is set to the smallest version that the
  changes has landed on
• All relevant documentation has been updated, for instance:
  • https://github.com/brave/brave-browser/wiki/Deviations-from-Chromium-(features-we-disable-or-remove)
  • https://github.com/brave/brave-browser/wiki/Proxy-redirected-URLs
  • https://github.com/brave/brave-browser/wiki/Fingerprinting-Protections
  • https://github.com/brave/brave-browser/wiki/Brave%E2%80%99s-Use-of-Referral-Codes
  • https://github.com/brave/brave-browser/wiki/Web-Compatibility-Exceptions-in-Brave
  • https://github.com/brave/brave-browser/wiki/QA-Guide
  • https://github.com/brave/brave-browser/wiki/P3A

Test Plan:

[tmancey]

Unit tests will fail (or they should do). Can you please update the unit tests too. Thanks

[tmancey]

@ptjames can you please add a link to what issue this resolves and a test plan if this requires QA, see description, thanks

[tmancey]

LGTM++

[tmancey]

@ptjames approved. Could you please raise an issue, link to it in the description, and add a test plan, before merging! Once merged, can you please add max_version to the Griffin seed for the associated experiment so that the experiment is removed after ~1 year. Thank you very much.

[tmancey]

@ptjames could you please squash any fix-up commits. Thanks

[tmancey]

LGTM++

[github-actions]

[puLL-Merge] - brave/brave-core@24253

Description

This PR makes several changes to the ad serving and prediction models in the Brave Ads system. The main changes include:

1. Updating the ad targeting logic to use a hierarchical segment structure.
2. Modifying the scoring system for ad relevance, particularly by reducing the weight of the "last seen ad" factor.
3. Adjusting unit tests to reflect these changes in targeting and scoring.

The motivation for these changes appears to be improving the relevance and effectiveness of ad targeting while simplifying some aspects of the prediction model.

Changes

Changes

1. eligible_inline_content_ads_v2_unittest.cc, eligible_new_tab_page_ads_v2_unittest.cc, eligible_notification_ads_v2_unittest.cc:

   • Updated test cases to use a hierarchical segment structure (e.g., "parent", "parent-child").
   • Modified expectations for ad selection based on the new targeting logic.

2. creative_ad_model_based_predictor_feature.h files:

   • Changed the default weight for the "last seen ad" predictor from 1.0 to 0.0 across different ad types.

3. creative_ad_model_based_predictor_last_seen_input_variable_info.h:

   • Changed the default weight for the "last seen ad" input variable from 1.0 to 0.0.

4. Various *_unittest.cc files:

   • Updated test expectations to reflect the changes in scoring and weighting, particularly the reduced influence of the "last seen ad" factor.

5. creative_ad_model_based_predictor_scoring_unittest.cc:

   • Adjusted expected scores based on the new weighting system.

Possible Issues

1. The significant reduction in the weight of the "last seen ad" factor (from 1.0 to 0.0) might lead to more frequent repetition of ads for users. This could potentially impact user experience if not carefully managed.

2. The changes to the targeting logic using hierarchical segments may require updates to the ad catalog or segmentation system to ensure compatibility.

Security Hotspots

No significant security issues were identified in this change. The modifications primarily affect internal ad serving logic and do not appear to introduce new attack vectors or expose sensitive information.