Q-LAtte: An Efficient and Versatile LSTM Model for Quantized Attention-Based Time Series Forecasting in Building Energy Applications

Computer Science

• Long Short-Term Memory Network 100%
• Memory Model 100%
• Decision-Making 100%
• Lstm 100%
• Approximation (Algorithm) 50%
• Computational Complexity 50%
• Real-Time Application 50%
• Superior Performance 50%
• Performance Model 50%
• Traditional Method 50%
• Attention (Machine Learning) 50%
• Constrained Device 50%
• Performance Degradation 50%
• Time Series Data 50%
• Quantization Process 50%

Engineering

• Energy Application 100%
• Long Short-Term Memory 100%
• Model Parameter 50%
• Performance Degradation 50%
• Data Series 50%
• Computational Resource 50%
• Computational Complexity 50%