Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911861974237184 |
|---|---|
| author | Lekeufack, Jordan Angelopoulos, Anastasios N. Bajcsy, Andrea Jordan, Michael I. Malik, Jitendra |
| author_facet | Lekeufack, Jordan Angelopoulos, Anastasios N. Bajcsy, Andrea Jordan, Michael I. Malik, Jitendra |
| contents | We introduce Conformal Decision Theory, a framework for producing safe autonomous decisions despite imperfect machine learning predictions. Examples of such decisions are ubiquitous, from robot planning algorithms that rely on pedestrian predictions, to calibrating autonomous manufacturing to exhibit high throughput and low error, to the choice of trusting a nominal policy versus switching to a safe backup policy at run-time. The decisions produced by our algorithms are safe in the sense that they come with provable statistical guarantees of having low risk without any assumptions on the world model whatsoever; the observations need not be I.I.D. and can even be adversarial. The theory extends results from conformal prediction to calibrate decisions directly, without requiring the construction of prediction sets. Experiments demonstrate the utility of our approach in robot motion planning around humans, automated stock trading, and robot manufacturing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_05921 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions Lekeufack, Jordan Angelopoulos, Anastasios N. Bajcsy, Andrea Jordan, Michael I. Malik, Jitendra Machine Learning Robotics Methodology We introduce Conformal Decision Theory, a framework for producing safe autonomous decisions despite imperfect machine learning predictions. Examples of such decisions are ubiquitous, from robot planning algorithms that rely on pedestrian predictions, to calibrating autonomous manufacturing to exhibit high throughput and low error, to the choice of trusting a nominal policy versus switching to a safe backup policy at run-time. The decisions produced by our algorithms are safe in the sense that they come with provable statistical guarantees of having low risk without any assumptions on the world model whatsoever; the observations need not be I.I.D. and can even be adversarial. The theory extends results from conformal prediction to calibrate decisions directly, without requiring the construction of prediction sets. Experiments demonstrate the utility of our approach in robot motion planning around humans, automated stock trading, and robot manufacturing. |
| title | Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions |
| topic | Machine Learning Robotics Methodology |
| url | https://arxiv.org/abs/2310.05921 |