Constructive Approach
Constructive approaches in machine learning focus on building models and algorithms to solve specific problems, often by integrating diverse data sources and leveraging pre-trained models for efficiency. Current research emphasizes the use of deep learning architectures, including convolutional neural networks and transformers, alongside techniques like ensemble learning, transfer learning, and meta-learning, to improve model performance and interpretability across various domains. These approaches are proving valuable in diverse applications, ranging from medical image analysis and fake news detection to robotics and space mission planning, demonstrating the broad impact of constructive methodologies on scientific advancement and practical problem-solving.
Papers
Revisiting MAB based approaches to recursive delegation
Nir Oren
Scalable Probabilistic Forecasting in Retail with Gradient Boosted Trees: A Practitioner's Approach
Xueying Long, Quang Bui, Grady Oktavian, Daniel F. Schmidt, Christoph Bergmeir, Rakshitha Godahewa, Seong Per Lee, Kaifeng Zhao, Paul Condylis
Bridging Distributionally Robust Learning and Offline RL: An Approach to Mitigate Distribution Shift and Partial Data Coverage
Kishan Panaganti, Zaiyan Xu, Dileep Kalathil, Mohammad Ghavamzadeh
An Approach to Automatically generating Riddles aiding Concept Attainment
Niharika Sri Parasa, Chaitali Diwan, Srinath Srinivasa
Bio-inspired computational memory model of the Hippocampus: an approach to a neuromorphic spike-based Content-Addressable Memory
Daniel Casanueva-Morato, Alvaro Ayuso-Martinez, Juan P. Dominguez-Morales, Angel Jimenez-Fernandez, Gabriel Jimenez-Moreno
LARA: A Light and Anti-overfitting Retraining Approach for Unsupervised Time Series Anomaly Detection
Feiyi Chen, Zhen Qin, Yingying Zhang, Shuiguang Deng, Yi Xiao, Guansong Pang, Qingsong Wen