Filters








1,478 Hits in 4.0 sec

A Conceptual Framework for Lifelong Learning [article]

Charles X. Ling, Tanner Bohn
2020 arXiv   pre-print
In this work, we propose a simple yet powerful unified framework that supports almost all of these properties and approaches through one central mechanism.  ...  Several lines of machine learning research, such as lifelong learning, few-shot learning, and transfer learning, attempt to capture these properties.  ...  A Unified Framework for Lifelong Learning In this section we describe how our unified framework works.  ... 
arXiv:1911.09704v4 fatcat:uwi5cntf2nb6posn3b2syppx5i

A Deep Learning Framework for Lifelong Machine Learning [article]

Charles X. Ling, Tanner Bohn
2021 arXiv   pre-print
In this work, we propose a simple yet powerful unified deep learning framework that supports almost all of these properties and approaches through one central mechanism.  ...  As academics, we often lack resources required to build and train, deep neural networks with billions of parameters on hundreds of TPUs.  ...  NSERC invests annually over $1 billion in people, discovery and innovation.  ... 
arXiv:2105.00157v1 fatcat:jnfc3jy7zjbovleubzu2vrfo7a

Enhancing Transferability of Black-box Adversarial Attacks via Lifelong Learning for Speech Emotion Recognition Models

Zhao Ren, Jing Han, Nicholas Cummins, Björn W. Schuller
2020 Zenodo  
A paper published in the INRTERSPEECH 2020 proceedings.  ...  A range of deep learning topologies have been successfully applied to the task of SER, such as Convolutional Neural Networks (CNNs) [5] , and Recurrent Neural Networks (RNNs) [6] .  ...  Conclusions and Future Work In summary, we trained an atrous Convolutional Neural Network (CNN) as a black-box adversarial attack model, and improved its transferability using lifelong learning.  ... 
doi:10.5281/zenodo.4251529 fatcat:2y7jzops7vf67g3gaotdxxgsx4

Towards Lifelong Self-Supervision: A Deep Learning Direction for Robotics [article]

Jay M. Wong
2016 arXiv   pre-print
These recent advances are only a piece to the puzzle. We suggest that deep learning as a tool alone is insufficient in building a unified framework to acquire general intelligence.  ...  Despite outstanding success in vision amongst other domains, many of the recent deep learning approaches have evident drawbacks for robots.  ...  Grupen for their expertise, enthusiasm, and insight on preliminary work with control state prediction networks that lead up to this direction of thinking.  ... 
arXiv:1611.00201v1 fatcat:nqatcsrysvd7nn3ljpc4eankmq

Analyzing Knowledge Transfer in Deep Q-Networks for Autonomously Handling Multiple Intersections [article]

David Isele, Akansel Cosgun, Kikuo Fujimura
2017 arXiv   pre-print
We view intersection handling as a deep reinforcement learning problem, which approximates the state action Q function as a deep neural network.  ...  Finally, we examine a lifelong learning setting, where we train a single network on five different types of intersections sequentially and show that the resulting network exhibited catastrophic forgetting  ...  reverse transfer and d) lifelong learning of multiple intersections with a single deep neural network.  ... 
arXiv:1705.01197v1 fatcat:5zswvdyrbzgc7jh7fantergvci

Representative Task Self-selection for Flexible Clustered Lifelong Learning [article]

Gan Sun, Yang Cong, Qianqian Wang, Bineng Zhong, Yun Fu
2019 arXiv   pre-print
However, the knowledge libraries or deep networks for most recent lifelong learning models are with prescribed size, and can degenerate the performance for both learned tasks and coming ones when facing  ...  Consider the lifelong machine learning paradigm whose objective is to learn a sequence of tasks depending on previous experiences, e.g., knowledge library or deep network weights.  ...  Eric Eaton for his constructive suggestion.  ... 
arXiv:1903.02173v2 fatcat:vrsfqiuowzfhflmeqtlr5y33uy

Embodied Artificial Intelligence through Distributed Adaptive Control: An Integrated Framework [article]

Clément Moulin-Frier, Jordi-Ysard Puigbò, Xerxes D. Arsiwalla, Martì Sanchez-Fibla, Paul F. M. J. Verschure
2017 arXiv   pre-print
Recurrent Neural Networks) with more traditional ones (e.g.  ...  Building upon this analysis, we first propose an embodied cognitive architecture integrating heterogenous sub-fields of Artificial Intelligence into a unified framework.  ...  We note that Deep Q-Learning [3] provides an integrated solution to the three processes involved in the Adaptive layer, based on Deep Convolutional Networks for Representation Learning, Q-value estimation  ... 
arXiv:1704.01407v3 fatcat:jqdvpkzjrnedzpxfz7bxhfn5mm

A Hybrid Temporal Data Mining Method for Intelligent Train Braking Systems

Wen Jing Liu, Guo Chun Wan, Mei Song Tong
2022 IEEE Access  
Then a predictive algorithm for model verification and update for lifelong learning is established to automatically update model parameters over time.  ...  This paper focuses on combining latest technology such as machine learning, transfer learning and lifelong learning to construct the first predictive analysis research framework in the field of train braking  ...  By taking advantages of both deep learning and optimal two-sample matching, a unified deep adaptation framework for jointly learning transferable representation and classifier is proposed to enable scalable  ... 
doi:10.1109/access.2022.3157598 fatcat:nwqunhfhb5gmxgwyiqnqcd6bxm

Towards Training Recurrent Neural Networks for Lifelong Learning [article]

Shagun Sodhani, Sarath Chandar, Yoshua Bengio
2019 arXiv   pre-print
In this work, we study these challenges in the context of sequential supervised learning with an emphasis on recurrent neural networks.  ...  Evaluation on the proposed benchmark shows that the unified model is more suitable than the constituent models for lifelong learning setting.  ...  Next, we describe how we extend the Net2Net model for RNNs. Then, we describe how the proposed model leverages both these mechanisms in a unified lifelong learning framework.  ... 
arXiv:1811.07017v3 fatcat:nzpp4kati5celf5f72u47plusu

Deep Learning in Mobile and Wireless Networking: A Survey [article]

Chaoyun Zhang, Paul Patras, Hamed Haddadi
2019 arXiv   pre-print
In this paper we bridge the gap between deep learning and mobile and wireless networking research, by presenting a comprehensive survey of the crossovers between the two areas.  ...  We first briefly introduce essential background and state-of-the-art in deep learning techniques with potential applications to networking.  ...  [66] A survey of deep learning in IoT data analytics. D D D Ahad et al. [67] A survey of neural networks in wireless networks. D D Mao et al.  ... 
arXiv:1803.04311v3 fatcat:awuvyviarvbr5kd5ilqndpfsde

Deep Learning in Mobile and Wireless Networking: A Survey

Chaoyun Zhang, Paul Patras, Hamed Haddadi
2019 IEEE Communications Surveys and Tutorials  
In this paper we bridge the gap between deep learning and mobile and wireless networking research, by presenting a comprehensive survey of the crossovers between the two areas.  ...  We first briefly introduce essential background and state-of-theart in deep learning techniques with potential applications to networking.  ...  [22] [69] A survey of neural networks in wireless networks. Mao et al. [70] A survey of deep learning for wireless networks. Luong et al.  ... 
doi:10.1109/comst.2019.2904897 fatcat:xmmrndjbsfdetpa5ef5e3v4xda

Knowledge Distillation: A Survey [article]

Jianping Gou, Baosheng Yu, Stephen John Maybank, Dacheng Tao
2021 arXiv   pre-print
In recent years, deep neural networks have been successful in both industry and academia, especially for computer vision tasks.  ...  As a representative type of model compression and acceleration, knowledge distillation effectively learns a small student model from a large teacher model.  ...  Lifelong Distillation Lifelong learning, including continual learning, continuous learning and meta-learning, aims to learn in a Low-precision student network Q u a n t i z a t i o n Full-precision teacher  ... 
arXiv:2006.05525v6 fatcat:aedzaeln5zf3jgjsgsn5kvjrri

Improved Schemes for Episodic Memory-based Lifelong Learning [article]

Yunhui Guo, Mingrui Liu, Tianbao Yang, Tajana Rosing
2020 arXiv   pre-print
Current deep neural networks can achieve remarkable performance on a single task.  ...  However, when the deep neural network is continually trained on a sequence of tasks, it seems to gradually forget the previous learned knowledge.  ...  Current deep neural networks are capable of achieving remarkable performance on a single task [6] .  ... 
arXiv:1909.11763v7 fatcat:7eg22i7bobchtlmm3ruut5ccsq

Continual Competitive Memory: A Neural System for Online Task-Free Lifelong Learning [article]

Alexander G. Ororbia
2021 arXiv   pre-print
In this article, we propose a novel form of unsupervised learning, continual competitive memory (CCM), as well as a computational framework to unify related neural models that operate under the principles  ...  The resulting neural system is shown to offer an effective approach for combating catastrophic forgetting in online continual classification problems.  ...  Under the problem context above, our contributions in this article are as follows: • We craft a simple modeling framework -the neural competitive learning (NCL) framework -for unifying neural models that  ... 
arXiv:2106.13300v1 fatcat:rl7n6riftbfztmbkrxvbzeutwq

Deep Learning for Visual SLAM in Transportation Robotics: A review

Chao Duan, Steffen Junginger, Jiahao Huang, Kairong Jin, Kerstin Thurow
2019 Transportation Safety and Environment  
With the great achievements of deep learning methods in the field of computer vision, there is a trend of applying deep learning methods to visual SLAM.  ...  In this paper, the latest research progress of deep learning applied to the field of visual SLAM is reviewed.  ...  A major challenge is to enable deep learning networks for lifelong learning visual SLAM systems.  ... 
doi:10.1093/tse/tdz019 fatcat:c5tj64xro5ftvcw6qwz7rgrgky
« Previous Showing results 1 — 15 out of 1,478 results