Hearing Device Learning Machine Adaptive Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hearing devices require active user feedback for training, which can be annoying and reduce user acceptance, and may not efficiently learn optimal settings over time.
Innovation Solution
A method that combines passive and experimentation-based training procedures for a learning machine within the hearing device, where settings are adjusted automatically or manually based on user feedback, allowing the device to learn optimal settings without frequent user interaction by dynamically updating parameter settings based on environmental situations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If active user feedback is required for training the learning machine, then the learning of optimal settings can be achieved, but user acceptance decreases and the operation becomes annoying
Solution Approach 1:
The learning machine performs self-training by automatically generating feedback signals based on its own performance. The system evaluates its classification accuracy and generates feedback when improvement opportunities are detected, eliminating the need for active user participation in the training process while maintaining learning effectiveness
Solution Approach 2:
The system implements an automatic feedback mechanism where the learning machine monitors its own classification performance and generates feedback signals internally. This feedback is used to retrain the learning machine and improve its classification accuracy without requiring external user input, thus resolving the contradiction between learning accuracy and user acceptance
2Productivity
If the learning machine is retrained frequently to improve classification accuracy, then optimal settings are learned faster, but more feedback signals are required which increases system complexity
Solution Approach 1:
The system performs preliminary evaluation of classification accuracy before initiating retraining. By assessing whether the current classification accuracy meets predefined criteria, the system avoids unnecessary retraining operations and the associated feedback generation, thus improving learning efficiency while reducing system complexity
Solution Approach 2:
The feedback generation mechanism is made dynamic by conditioning feedback creation on actual classification accuracy measurements. The system adaptively determines when feedback is needed based on performance metrics, allowing flexible retraining frequency that optimizes learning speed while minimizing the complexity of the feedback mechanism
Data Source
AI summary
A hearing device has a signal processor which has an adjustable parameter that has a given setting at a given time. The parameter is set depending on the situation by selecting a setting, depending on an environmental situation and by a learning machine. A current setting of the parameter can be rated by feedback from a user. In a first training procedure the learning machine is passively trained by negative feedback signals, by rating feedback from the user as dissatisfaction with the current setting and by assuming the user's satisfaction with the current setting as long as no feedback is given. In a second training procedure the learning machine is trained by changing the current setting independently of the feedback from the user and in spite of an assumed satisfaction with the current setting, so that the user is offered a different setting which can then be rated by feedback.