Expert system gradient descent style training: Development of a defensible artificial intelligence technique

نویسندگان

چکیده

Artificial intelligence systems, which are designed with a capability to learn from the data presented them, used throughout society. These systems screen loan applicants, make sentencing recommendations for criminal defendants, scan social media posts disallowed content and more. Because these don't assign meaning their complex learned correlation network, they can associations that equate causality, resulting in non-optimal indefensible decisions being made. In addition making sub-optimal, may create legal liability designers operators by learning correlations violate anti-discrimination other laws regarding what factors be different types of decision making. This paper presents use machine expert system, is developed meaning-assigned nodes (facts) (rules). Multiple potential implementations considered evaluated under conditions, including network error augmentation levels training levels. The performance compared random fully connected networks.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Handwritten Character Recognition using Modified Gradient Descent Technique of Neural Networks and Representation of Conjugate Descent for Training Patterns

The purpose of this study is to analyze the performance of Back propagation algorithm with changing training patterns and the second momentum term in feed forward neural networks. This analysis is conducted on 250 different words of three small letters from the English alphabet. These words are presented to two vertical segmentation programs which are designed in MATLAB and based on portions (1...

متن کامل

Gradient Descent and Fast Artificial Time Integration

The integration to steady state of many initial value ODEs and PDEs using the forward Euler method can alternatively be considered as gradient descent for an associated minimization problem. Greedy algorithms such as steepest descent for determining the step size are as slow to reach steady state as is forward Euler integration with the best uniform step size. But other, much faster methods usi...

متن کامل

Expert System for Coronary Heart Disease - Built Using Artificial Intelligence

Coronary Heart Disease is a disease which is difficult to diagnose and is vey commonly identified only during the mortality of an individual. The World Health Organization (WHO) reported that 70 per cent deaths occur in subjects less than 70 years of age in India and in other developing countries. Since Coronary Heart Disease (CHD) is becoming an epidemic in India, there is a terrific need for ...

متن کامل

Gradient Descent Training of Bayesian Networks

As shown by Russel et al., 1995 7], Bayesian networks can be equipped with a gradient descent learning method similar to the training method for neural networks. The calculation of the required gradients can be performed locally along with propagation. We review how this can be done, and we show how the gradient descent approach can be used for various tasks like tuning and training with traini...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Knowledge Based Systems

سال: 2021

ISSN: ['1872-7409', '0950-7051']

DOI: https://doi.org/10.1016/j.knosys.2021.107275