In recent decades, advancements in information technology allowed Artificial Intelligence (AI) systems to predict fu- ture outcomes with unprecedented success. This brought the widespread deployment of these methods in many fields, in- tending to support decision-making. A pressing question is how to make AI systems trustworthy and robust to common challenges in real-life scenarios. In my work, I plan to ex- plore ways to enhance the trustworthiness of AI through the selective classification framework. In this setting, the AI sys- tem can refrain from predicting whenever it is not confident enough, allowing it to trade off coverage, i.e. the percentage of instances that receive a prediction, for performance.
Topics in Selective Classification
Pugnana, Andrea
2023
Abstract
In recent decades, advancements in information technology allowed Artificial Intelligence (AI) systems to predict fu- ture outcomes with unprecedented success. This brought the widespread deployment of these methods in many fields, in- tending to support decision-making. A pressing question is how to make AI systems trustworthy and robust to common challenges in real-life scenarios. In my work, I plan to ex- plore ways to enhance the trustworthiness of AI through the selective classification framework. In this setting, the AI sys- tem can refrain from predicting whenever it is not confident enough, allowing it to trade off coverage, i.e. the percentage of instances that receive a prediction, for performance.File | Dimensione | Formato | |
---|---|---|---|
TopicsInSelectiveClassification.pdf
accesso aperto
Tipologia:
Published version
Licenza:
Solo Lettura
Dimensione
54.83 kB
Formato
Adobe PDF
|
54.83 kB | Adobe PDF |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.