.. _sphx_glr_examples_risk_control_2-advanced-analysis:

.. _risk_control_examples_2:

2. Advanced analysis
--------------------

The following examples use MAPIE for discussing more complex risk control problems.


.. raw:: html

    <div class="sphx-glr-thumbnails">

.. thumbnail-parent-div-open

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" In this example, we explain how to perform risk control for multi-label classification using the Learn-Then-Test (LTT) procedure implemented in MAPIE.">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_risk_control_multi-label_classification_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_risk_control_multi-label_classification.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Control the risk of a multi-label classifier</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" AI is a powerful tool for email sorting (for example between spam and urgent emails). However, because algorithms are not perfect, manual verification is sometimes required. Thus one would like to be able to control the amount of emails sent to human validation. One way to do so is to define a multi-parameter prediction function based on a classifier&#x27;s predicted scores. This would allow defining a rule for email checking, which could be adapted by varying the prediction parameters.">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_risk_control_multi_parameter_binary_classification_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_risk_control_multi_parameter_binary_classification.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Control risk of a binary classifier with multiple prediction parameters</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" This example illustrates how to control the recall of a semantic segmentation model using MAPIE.">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_semantic_segmentation_recall_control_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_semantic_segmentation_recall_control.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Recall control for semantic segmentation</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" This example illustrates how to control the precision of a semantic segmentation model using MAPIE.">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_semantic_segmentation_precision_control_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_semantic_segmentation_precision_control.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Precision control for semantic segmentation</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" This example compares how different family-wise error rate (FWER) control strategies affect the set of statistically valid thresholds when controlling a risk in binary classification.">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_fwer_method_comparison_for_recall_guarantees_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_fwer_method_comparison_for_recall_guarantees.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Comparing FWER methods for risk control in binary classification</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" This example demonstrates how to control a non-monotonic risk such as 1 − precision in binary classification using multiple family-wise error rate (FWER) procedures, with a particular focus on Split Fixed Sequence Testing (SFST).">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_split_fixed_sequence_testing_fwer_method_for_precision_control_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_split_fixed_sequence_testing_fwer_method_for_precision_control.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Split Fixed Sequence Testing for Precision Control under Multiple Testing</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip=" In this example, we explain how to do multi-risk control for binary classification with MAPIE.">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_multi_risk_control_binary_classification_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_multi_risk_control_binary_classification.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Control multiple risks of a binary classifier</div>
    </div>


.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="This example demonstrates how to use risk control methods for Large Language Models (LLMs) acting as judges. We simulate a scenario where an LLM evaluates answers, and we want to control the risk of hallucination detection. Moreover, we want the judge to abstain from making a decision when it is uncertain.">

.. only:: html

  .. image:: /examples_risk_control/2-advanced-analysis/images/thumb/sphx_glr_plot_risk_control_llm_as_a_judge_thumb.png
    :alt:

  :ref:`sphx_glr_examples_risk_control_2-advanced-analysis_plot_risk_control_llm_as_a_judge.py`

.. raw:: html

      <div class="sphx-glr-thumbnail-title">Risk Control for LLM as a Judge with Abstention</div>
    </div>


.. thumbnail-parent-div-close

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /examples_risk_control/2-advanced-analysis/plot_risk_control_multi-label_classification
   /examples_risk_control/2-advanced-analysis/plot_risk_control_multi_parameter_binary_classification
   /examples_risk_control/2-advanced-analysis/plot_semantic_segmentation_recall_control
   /examples_risk_control/2-advanced-analysis/plot_semantic_segmentation_precision_control
   /examples_risk_control/2-advanced-analysis/plot_fwer_method_comparison_for_recall_guarantees
   /examples_risk_control/2-advanced-analysis/plot_split_fixed_sequence_testing_fwer_method_for_precision_control
   /examples_risk_control/2-advanced-analysis/plot_multi_risk_control_binary_classification
   /examples_risk_control/2-advanced-analysis/plot_risk_control_llm_as_a_judge