NEMO : Neuro-Evolution with Multiobjective Optimization of Deep Neural Network for Speed and Accuracy
By Ye-Hoon Kim, B. Reddy, Sojung Yun, Chanwon Seo. 2017
80 citations recorded by the source corpus.
Subfields: Neural Architecture Search & AutoML, Evolutionary Computation
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Neural Architecture Search & AutoML
- Xiangxiang Chu
- Ruijun Xu
- Frank Hutter
- Martin Wistuba
- Thomas Elsken
- Tejaswini Pedapati
- Qi Tian
- Jan Hendrik Metzen
In Evolutionary Computation
Citations in the corpus
- Efficient Multi-Objective Neural Architecture Search via Lamarckian Evolution (2018, background): “Kim et al. (2017) parameterize an architecture by a fixed-length vector describing, which limits the architecture search space drastically.”
- A Survey on Neural Architecture Search (2019, methodology): “Notable works include methods that attempt to limit the number of model parameters or the like, for efficient deployment on mobile devices (Tan et al., 2018; Kim et al., 2017).”
- NSGA-NET: A Multi-Objective Genetic Algorithm for Neural Architecture Search (2018, methodology): “Kim et al. Kim et al. (2017) present an algorithm utilizing NSGA-II Deb et al. (2000), however their method only searches over hyperparameters and a small fixed set of architectures.”
- DPP-Net: Device-aware Progressive Search for Pareto-optimal Neural Architectures (2018, background): “[19] proposed to treat neural network architecture search as a multi-objective optimization task and adopt an evolutionary algorithm to search models with two objectives, run-time speed, and classification accuracy.”
- Multi-Objective Reinforced Evolution in Mobile Neural Architecture Search (2019, methodology): “NEMO (Kim et al., 2017) and NSGA-Net (Lu et al., 2018) adopt the classic non-dominated search algorithm (NSGA-II) to handle trade-off among multi-objectives.”
- PPP-Net: Platform-aware Progressive Search for Pareto-optimal Neural Architectures (2018, methodology): “Inspired by Liu et al. (2017a), we adopt Sequential Model-Based Optimization (Hutter et al.”
- NSGA-Net: neural architecture search using multi-objective genetic algorithm (2018, unknown): “Kim et al. [21] presents an algo-rithm utilizing NSGA-II [5], however their method only searches over hyper-parameters and a small fixed set of architectures.”
- Hardware/Software Co-Exploration of Neural Architectures (2019, unknown): “In the existing work, there are two mainstreams of architecture search: 1) employing reinforcement learning [2], [16], [26] and 2) applying evolutionary algorithms [3], [27], [28].”
- CARS: Continuous Evolution for Efficient Neural Architecture Search (2019, background): “NEMO [23] and MNasNet [41] target at speed and accuracy.”
- InstaNAS: Instance-aware Neural Architecture Search (2018, background): “Elsken et al. [6] and Zhou et al. [31] use FLOPs and the number of parameters as the proxies of computational costs; Kim et al. [9] and Tan et al. [24] directly minimized the actual inference time; Dong et al. [5] proposed to consider both device-agnostic objectives ( e.g ., FLOPs) and…”