Real-Time Low-Cost Drift Compensation for Chemical Sensors Using a Deep Neural Network with Hadamard Transform and Additive Layers

Engineering & Materials Science

• Hadamard transforms 100%
• Chemical sensors 78%
• Deep neural networks 67%
• Compensation and Redress 53%
• Discrete cosine transforms 39%
• Costs 26%
• Neural networks 21%
• Convolution 17%
• Deep learning 15%
• Sensors 8%

Physics & Astronomy

• discrete cosine transform 43%
• sensors 42%
• multiplication 28%
• learning 15%
• convolution integrals 14%
• central processing units 14%
• sensitivity 8%
• estimates 8%