-
A Principled Approach to Failure Analysis and Model Repairment: Demonstration in Medical Imaging
Authors:
Thomas Henn,
Yasukazu Sakamoto,
Clément Jacquet,
Shunsuke Yoshizawa,
Masamichi Andou,
Stephen Tchen,
Ryosuke Saga,
Hiroyuki Ishihara,
Katsuhiko Shimizu,
Yingzhen Li,
Ryutaro Tanno
Abstract:
Machine learning models commonly exhibit unexpected failures post-deployment due to either data shifts or uncommon situations in the training environment. Domain experts typically go through the tedious process of inspecting the failure cases manually, identifying failure modes and then attempting to fix the model. In this work, we aim to standardise and bring principles to this process through an…
▽ More
Machine learning models commonly exhibit unexpected failures post-deployment due to either data shifts or uncommon situations in the training environment. Domain experts typically go through the tedious process of inspecting the failure cases manually, identifying failure modes and then attempting to fix the model. In this work, we aim to standardise and bring principles to this process through answering two critical questions: (i) how do we know that we have identified meaningful and distinct failure types?; (ii) how can we validate that a model has, indeed, been repaired? We suggest that the quality of the identified failure types can be validated through measuring the intra- and inter-type generalisation after fine-tuning and introduce metrics to compare different subty** methods. Furthermore, we argue that a model can be considered repaired if it achieves high accuracy on the failure types while retaining performance on the previously correct data. We combine these two ideas into a principled framework for evaluating the quality of both the identified failure subtypes and model repairment. We evaluate its utility on a classification and an object detection tasks. Our code is available at https://github.com/Rokken-lab6/Failure-Analysis-and-Model-Repairment
△ Less
Submitted 25 September, 2021;
originally announced September 2021.
-
Characterization of spin-orbit fields in InGaAs quantum wells
Authors:
T. Henn,
L. Czornomaz,
G. Salis
Abstract:
Coherent electron spin dynamics in 10-nm-wide InGaAs/InAlAs quantum wells is studied from 10 K to room temperature using time-resolved Kerr rotation. The spin lifetime exceeds 1 ns at 10 K and decreases with temperature. By varying the spatial overlap between pump and probe pulses, a diffusive velocity is imprinted on the measured electron spins and a spin precession in the spin-orbit field is mea…
▽ More
Coherent electron spin dynamics in 10-nm-wide InGaAs/InAlAs quantum wells is studied from 10 K to room temperature using time-resolved Kerr rotation. The spin lifetime exceeds 1 ns at 10 K and decreases with temperature. By varying the spatial overlap between pump and probe pulses, a diffusive velocity is imprinted on the measured electron spins and a spin precession in the spin-orbit field is measured. A Rashba symmetry of the SOI is determined. By comparing the spatial precession frequency gradient with the spin decay rate, an upper limit for the Rashba coefficients $α$ of 2$\times$10$^{-12}$ eVm is estimated.
△ Less
Submitted 19 September, 2016;
originally announced September 2016.
-
Selecting the Best Player Formation for Corner-Kick Situations Based on Bayes' Estimation
Authors:
Jordan Henrio,
Thomas Henn,
Tomoharu Nakashima,
Hidehisa Akiyama
Abstract:
In the domain of the Soccer simulation 2D league of the RoboCup project, appropriate player positioning against a given opponent team is an important factor of soccer team performance. This work proposes a model which decides the strategy that should be applied regarding a particular opponent team. This task can be realized by applying preliminary a learning phase where the model determines the mo…
▽ More
In the domain of the Soccer simulation 2D league of the RoboCup project, appropriate player positioning against a given opponent team is an important factor of soccer team performance. This work proposes a model which decides the strategy that should be applied regarding a particular opponent team. This task can be realized by applying preliminary a learning phase where the model determines the most effective strategies against clusters of opponent teams. The model determines the best strategies by using sequential Bayes' estimators. As a first trial of the system, the proposed model is used to determine the association of player formations against opponent teams in the particular situation of corner-kick. The implemented model shows satisfying abilities to compare player formations that are similar to each other in terms of performance and determines the right ranking even by running a decent number of simulation games.
△ Less
Submitted 3 June, 2016;
originally announced June 2016.
-
Ultrafast supercontinuum fiber-laser based pump-probe scanning MOKE microscope for the investigation of electron spin dynamics in semiconductors at cryogenic temperatures with picosecond time and micrometer spatial resolution
Authors:
T. Henn,
T. Kiessling,
W. Ossau,
L. W. Molenkamp,
K. Biermann,
P. V. Santos
Abstract:
We describe a two-color pump-probe scanning magneto-optical Kerr effect (MOKE) microscope which we have developed to investigate electron spin phenomena in semiconductors at cryogenic temperatures with picosecond time and micrometer spatial resolution. The key innovation of our microscope is the usage of an ultrafast `white light' supercontinuum fiber-laser source which provides access to the whol…
▽ More
We describe a two-color pump-probe scanning magneto-optical Kerr effect (MOKE) microscope which we have developed to investigate electron spin phenomena in semiconductors at cryogenic temperatures with picosecond time and micrometer spatial resolution. The key innovation of our microscope is the usage of an ultrafast `white light' supercontinuum fiber-laser source which provides access to the whole visible and near-infrared spectral range. Our Kerr microscope allows for the independent selection of the excitation and detection energy while avoiding the necessity to synchronize the pulse trains of two separate picosecond laser systems. The ability to independently tune the pump and probe wavelength enables the investigation of the influence of excitation energy on the optically induced electron spin dynamics in semiconductors. We demonstrate picosecond real-space imaging of the diffusive expansion of optically excited electron spin packets in a (110) GaAs quantum well sample to illustrate the capabilities of the instrument.
△ Less
Submitted 11 October, 2013;
originally announced October 2013.
