Zobrazeno 1 - 10
of 32
pro vyhledávání: '"Zela, Arber"'
Autor:
Sukthanker, Rhea Sanjay, Zela, Arber, Staffler, Benedikt, Klein, Aaron, Purucker, Lennart, Franke, Joerg K. H., Hutter, Frank
The increasing size of language models necessitates a thorough analysis across multiple dimensions to assess trade-offs among crucial hardware metrics such as latency, energy consumption, GPU memory usage, and performance. Identifying optimal model c
Externí odkaz:
http://arxiv.org/abs/2405.10299
Autor:
Sukthanker, Rhea Sanjay, Zela, Arber, Staffler, Benedikt, Dooley, Samuel, Grabocka, Josif, Hutter, Frank
Pareto front profiling in multi-objective optimization (MOO), i.e. finding a diverse set of Pareto optimal solutions, is challenging, especially with expensive objectives like neural network training. Typically, in MOO neural architecture search (NAS
Externí odkaz:
http://arxiv.org/abs/2402.18213
Autor:
White, Colin, Safari, Mahmoud, Sukthanker, Rhea, Ru, Binxin, Elsken, Thomas, Zela, Arber, Dey, Debadeepta, Hutter, Frank
In the past decade, advances in deep learning have resulted in breakthroughs in a variety of areas, including computer vision, natural language understanding, speech recognition, and reinforcement learning. Specialized, high-performing neural archite
Externí odkaz:
http://arxiv.org/abs/2301.08727
Zero-cost proxies (ZC proxies) are a recent architecture performance prediction technique aiming to significantly speed up algorithms for neural architecture search (NAS). Recent work has shown that these techniques show great promise, but certain as
Externí odkaz:
http://arxiv.org/abs/2210.03230
Autor:
Elsken, Thomas, Zela, Arber, Metzen, Jan Hendrik, Staffler, Benedikt, Brox, Thomas, Valada, Abhinav, Hutter, Frank
The success of deep learning in recent years has lead to a rising demand for neural network architecture engineering. As a consequence, neural architecture search (NAS), which aims at automatically designing neural network architectures in a data-dri
Externí odkaz:
http://arxiv.org/abs/2202.07242
Autor:
Mehta, Yash, White, Colin, Zela, Arber, Krishnakumar, Arjun, Zabergja, Guri, Moradian, Shakiba, Safari, Mahmoud, Yu, Kaicheng, Hutter, Frank
The release of tabular benchmarks, such as NAS-Bench-101 and NAS-Bench-201, has significantly lowered the computational overhead for conducting scientific research in neural architecture search (NAS). Although they have been widely adopted and used t
Externí odkaz:
http://arxiv.org/abs/2201.13396
Ensembles of CNN models trained with different seeds (also known as Deep Ensembles) are known to achieve superior performance over a single copy of the CNN. Neural Ensemble Search (NES) can further boost performance by adding architectural diversity.
Externí odkaz:
http://arxiv.org/abs/2107.04369
While neural architecture search methods have been successful in previous years and led to new state-of-the-art performance on various problems, they have also been criticized for being unstable, being highly sensitive with respect to their hyperpara
Externí odkaz:
http://arxiv.org/abs/2107.03719
Early methods in the rapidly developing field of neural architecture search (NAS) required fully training thousands of neural networks. To reduce this extreme computational cost, dozens of techniques have since been proposed to predict the final perf
Externí odkaz:
http://arxiv.org/abs/2104.01177
Neural architecture search (NAS) has recently been addressed from various directions, including discrete, sampling-based methods and efficient differentiable approaches. While the former are notoriously expensive, the latter suffer from imposing stro
Externí odkaz:
http://arxiv.org/abs/2010.04683