Difference between revisions of "Lifelong Learning"
m (BPeat moved page Catastrophic Forgetting Challenge to Lifelong Learning - Catastrophic Forgetting Challenge) |
|||
| Line 8: | Line 8: | ||
[http://www.google.com/search?q=Multi-model+Catastrophic+Forgetting ...Google search] | [http://www.google.com/search?q=Multi-model+Catastrophic+Forgetting ...Google search] | ||
| + | * [http://www.darpa.mil/program/lifelong-learning-machines Lifelong Learning Machines (L2M) | DARPA] | ||
| + | * [http://www.darpa.mil/news-events/2019-03-12 Progress on Lifelong Learning Machines Shows Potential for Bio-Inspired Algorithms | USC & DARPA] | ||
| + | * [http://www.darpa.mil/news-events/2018-05-03 Researchers Selected to Develop Novel Approaches to Lifelong Machine Learning | DARPA] | ||
* [[Automated Machine Learning (AML) - AutoML]] | * [[Automated Machine Learning (AML) - AutoML]] | ||
| − | In recent years, researchers have developed deep neural networks that can perform a variety of tasks, including visual recognition and natural language processing (NLP) tasks. Although many of these models achieved remarkable results, they typically only perform well on one particular task due to what is referred to as "catastrophic forgetting." Essentially, catastrophic forgetting means that when a model that was initially trained on task A is later trained on task B, its performance on task A will significantly decline. [http://techxplore.com/news/2019-03-approach-multi-model-deep-neural-networks.html A new approach to overcome multi-model forgetting in deep neural networks | Ingrid Fadelli | + | In recent years, researchers have developed deep neural networks that can perform a variety of tasks, including visual recognition and natural language processing (NLP) tasks. Although many of these models achieved remarkable results, they typically only perform well on one particular task due to what is referred to as "catastrophic forgetting." Essentially, catastrophic forgetting means that when a model that was initially trained on task A is later trained on task B, its performance on task A will significantly decline. [http://techxplore.com/news/2019-03-approach-multi-model-deep-neural-networks.html A new approach to overcome multi-model forgetting in deep neural networks] and [https://techxplore.com/news/2019-03-memory-approach-enable-lifelong.html A generative memory approach to enable lifelong reinforcement learning] | | Ingrid Fadelli |
http://3c1703fe8d.site.internapcdn.net/newman/csz/news/800/2019/11-anewapproach.jpg | http://3c1703fe8d.site.internapcdn.net/newman/csz/news/800/2019/11-anewapproach.jpg | ||
| Line 20: | Line 23: | ||
<youtube>OBkruvl8ih8</youtube> | <youtube>OBkruvl8ih8</youtube> | ||
<youtube>q4A76i6TOCc</youtube> | <youtube>q4A76i6TOCc</youtube> | ||
| + | |||
| + | * [http://techxplore.com/news/2018-10-developmental-approach-sensorimotor-space-enlargement.html A new developmental reinforcement learning approach for sensorimotor space enlargement | Ingrid Fadelli] | ||
| + | http://3c1703fe8d.site.internapcdn.net/newman/csz/news/800/2018/4-anewdevelopm.jpg | ||
Revision as of 05:25, 14 March 2019
YouTube search... ...Google search
- Lifelong Learning Machines (L2M) | DARPA
- Progress on Lifelong Learning Machines Shows Potential for Bio-Inspired Algorithms | USC & DARPA
- Researchers Selected to Develop Novel Approaches to Lifelong Machine Learning | DARPA
- Automated Machine Learning (AML) - AutoML
In recent years, researchers have developed deep neural networks that can perform a variety of tasks, including visual recognition and natural language processing (NLP) tasks. Although many of these models achieved remarkable results, they typically only perform well on one particular task due to what is referred to as "catastrophic forgetting." Essentially, catastrophic forgetting means that when a model that was initially trained on task A is later trained on task B, its performance on task A will significantly decline. A new approach to overcome multi-model forgetting in deep neural networks and A generative memory approach to enable lifelong reinforcement learning | | Ingrid Fadelli
