Zobrazeno 1 - 10
of 28
pro vyhledávání: '"Real, Esteban"'
How to automatically design better machine learning programs is an open problem within AutoML. While evolution has been a popular tool to search for better ML programs, using learning itself to guide the search has been less successful and less under
Externí odkaz:
http://arxiv.org/abs/2402.05821
Autor:
Real, Esteban, Chen, Yao, Rossini, Mirko, de Souza, Connal, Garg, Manav, Verghese, Akhil, Firsching, Moritz, Le, Quoc V., Cubuk, Ekin Dogus, Park, David H.
Computers calculate transcendental functions by approximating them through the composition of a few limited-precision instructions. For example, an exponential can be calculated with a Taylor series. These approximation methods were developed over th
Externí odkaz:
http://arxiv.org/abs/2312.08472
Autor:
Kelly, Stephen, Park, Daniel S., Song, Xingyou, McIntire, Mitchell, Nashikkar, Pranav, Guha, Ritam, Banzhaf, Wolfgang, Deb, Kalyanmoy, Boddeti, Vishnu Naresh, Tan, Jie, Real, Esteban
Autonomous robots deployed in the real world will need control policies that rapidly adapt to environmental changes. To this end, we propose AutoRobotics-Zero (ARZ), a method based on AutoML-Zero that discovers zero-shot adaptable policies from scrat
Externí odkaz:
http://arxiv.org/abs/2307.16890
Autor:
Chen, Xiangning, Liang, Chen, Huang, Da, Real, Esteban, Wang, Kaiyuan, Liu, Yao, Pham, Hieu, Dong, Xuanyi, Luong, Thang, Hsieh, Cho-Jui, Lu, Yifeng, Le, Quoc V.
We present a method to formulate algorithm discovery as program search, and apply it to discover optimization algorithms for deep neural network training. We leverage efficient search techniques to explore an infinite and sparse program space. To bri
Externí odkaz:
http://arxiv.org/abs/2302.06675
Autor:
Gillard, Ryan, Jonany, Stephen, Miao, Yingjie, Munn, Michael, de Souza, Connal, Dungay, Jonathan, Liang, Chen, So, David R., Le, Quoc V., Real, Esteban
The field of Automatic Machine Learning (AutoML) has recently attained impressive results, including the discovery of state-of-the-art machine learning solutions, such as neural image classifiers. This is often done by applying an evolutionary search
Externí odkaz:
http://arxiv.org/abs/2302.05433
The increasing complexity and scale of machine learning (ML) has led to the need for more efficient collaboration among multiple teams. For example, when a research team invents a new architecture like "ResNet," it is desirable for multiple engineeri
Externí odkaz:
http://arxiv.org/abs/2302.01918
Autor:
Garau-Luis, Juan Jose, Miao, Yingjie, Co-Reyes, John D., Parisi, Aaron, Tan, Jie, Real, Esteban, Faust, Aleksandra
Generalizability and stability are two key objectives for operating reinforcement learning (RL) agents in the real world. Designing RL algorithms that optimize these objectives can be a costly and painstaking process. This paper presents MetaPG, an e
Externí odkaz:
http://arxiv.org/abs/2204.04292
Autor:
Peng, Daiyi, Dong, Xuanyi, Real, Esteban, Tan, Mingxing, Lu, Yifeng, Liu, Hanxiao, Bender, Gabriel, Kraft, Adam, Liang, Chen, Le, Quoc V.
Neural networks are sensitive to hyper-parameter and architecture choices. Automated Machine Learning (AutoML) is a promising paradigm for automating these choices. Current ML software libraries, however, are quite limited in handling the dynamic int
Externí odkaz:
http://arxiv.org/abs/2101.08809
Autor:
Co-Reyes, John D., Miao, Yingjie, Peng, Daiyi, Real, Esteban, Levine, Sergey, Le, Quoc V., Lee, Honglak, Faust, Aleksandra
We propose a method for meta-learning reinforcement learning algorithms by searching over the space of computational graphs which compute the loss function for a value-based model-free RL agent to optimize. The learned algorithms are domain-agnostic
Externí odkaz:
http://arxiv.org/abs/2101.03958
Machine learning research has advanced in multiple aspects, including model structures and learning methods. The effort to automate such research, known as AutoML, has also made significant progress. However, this progress has largely focused on the
Externí odkaz:
http://arxiv.org/abs/2003.03384