Invention Application
- Patent Title: METHODS AND APPARATUS TO CALIBRATE ERROR ALIGNED UNCERTAINTY FOR REGRESSION AND CONTINUOUS STRUCTURED PREDICTION TASKS
-
Application No.: US17855774Application Date: 2022-06-30
-
Publication No.: US20220343171A1Publication Date: 2022-10-27
- Inventor: Neslihan Kose Cihangir , Omesh Tickoo , Ranganath Krishnan , Ignacio J. Alvarez , Michael Paulitsch , Akash Dhamasia
- Applicant: Neslihan Kose Cihangir , Omesh Tickoo , Ranganath Krishnan , Ignacio J. Alvarez , Michael Paulitsch , Akash Dhamasia
- Applicant Address: DE Munich; US OR Portland; US OR Hillsboro; US OR Portland; DE Ottobrunn; DE Munich
- Assignee: Neslihan Kose Cihangir,Omesh Tickoo,Ranganath Krishnan,Ignacio J. Alvarez,Michael Paulitsch,Akash Dhamasia
- Current Assignee: Neslihan Kose Cihangir,Omesh Tickoo,Ranganath Krishnan,Ignacio J. Alvarez,Michael Paulitsch,Akash Dhamasia
- Current Assignee Address: DE Munich; US OR Portland; US OR Hillsboro; US OR Portland; DE Ottobrunn; DE Munich
- Main IPC: G06N3/08
- IPC: G06N3/08

Abstract:
Methods, apparatus, systems, and articles of manufacture are disclosed that calibrate error aligned uncertainty for regression and continuous structured prediction tasks/optimizations. An example apparatus includes a prediction model, at least one memory, instructions, and processor circuitry to at least one of execute or instantiate the instructions to calculate a count of samples corresponding to an accuracy-certainty classification category, calculate a trainable uncertainty calibration loss value based on the calculated count, calculate a final differentiable loss value based on the trainable uncertainty calibration loss value, and calibrate the prediction model with the final differentiable loss value.
Information query