December 18, 2018
Deep learning offers the promise of bypassing the process of manual feature engineering by learning representations in conjunction with statistical models in an end-to-end fashion. However, neural network architectures themselves are typically designed by experts in a painstaking, ad hoc fashion. Neural architecture search (NAS) has been touted as the path forward for alleviating this pain by automatically identifying architectures that are superior to hand-designed ones.
But with the field moving so fast both in terms of research progress and hype, it can be hard to get answers to basic questions: What exactly is NAS and is it fundamentally different from AutoML or Hyperparameter Optimization? Do specialized NAS methods actually work? Aren’t they prohibitively expensive to use? Should I be using specialized NAS methods? In this post we’ll answer each of these questions. Our discussion touches upon a few key points:
To set the stage, let’s first discuss how NAS fits within the wider umbrella of AutoML (“automated machine learning”).
AutoML focuses on automating every aspect of the machine learning workflow to increase efficiency and democratize machine learning so that non-experts can apply machine learning to their problems with ease. While AutoML encompasses the automation of a wide range of problems associated with ETL, model training, and model deployment, the problem of Hyperparameter Optimization is a core focus of AutoML. This problem involves configuring the internal settings that govern the behavior of an ML model/algorithm in order to return a high-quality predictive model.
For example, ridge regression models require setting the value of a regularization term; random forest models require the user to set the maximum tree depth and minimum number of samples per leaf; and training any model with stochastic gradient descent requires setting an appropriate step size. Neural networks also require setting a multitude of hyperparameters including (1) selecting an optimization method along with its associated set of hyperparameters, (2) setting the dropout rate and other regularization hyperparameters, and, if desired, (3) tuning parameters that control the architecture of the network (e.g., number of hidden layers, number of convolutional filters).
Although the exposition on Neural Architecture Search (NAS) might suggest that it is a completely new problem, our final example above hints at a close relationship between hyperparameter optimization and NAS. While the search spaces used for NAS are generally larger and control different aspects of the neural network architecture, the underlying problem is the same as that addressed by hyperparameter optimization: find a configuration within the search space that performs well on the target task. Hence, we view NAS to be a subproblem within hyperparameter optimization.
NAS is nonetheless an exciting direction to study, as focusing on a specialized subproblem provides the opportunity to exploit additional structure to design custom tailored solutions, as is done by many specialized NAS approaches. In the next section, we will provide an overview of NAS and delve more into the similarities and differences between hyperparameter optimization and NAS.
Interest in NAS ballooned after the work of Zoph et. al. 2016 used reinforcement learning to design, at the time, state-of-the-art architectures for image recognition and language modeling. However, Zoph et. al. 2016, in addition to other first generation specialized approaches for NAS, required a tremendous amount of computational power (e.g., hundreds of GPUs running for thousands (!) of GPU days in aggregate), making them impractical for all but the likes of companies like Google. More recent approaches exploit various methods of reuse to drastically reduce the computational cost, and new methods are being rapidly introduced in the research community.
We’ll next dive a bit deeper into the core design decisions associated with all of these specialized NAS methods.2 The three main components are:
Notably, these are the same three requisite ingredients for traditional hyperparameter optimization methods. The research community has converged on a few canonical benchmarking datasets and tasks to evaluate the performance of different search methods, and we’ll next use these benchmarks to report results on head-to-head comparisons between (1) human designed architectures tuned via hyperparameter optimization methods, and (2) NAS designed architectures identified3 via leading specialized NAS methods.
The two most common tasks used to benchmark NAS methods are (1) designing CNN architectures evaluated on the CIFAR-10 dataset, and (2) designing RNN architectures evaluated on the PennTree Bank (PTB) dataset. We show the test error for different architectures on CIFAR-10 in the table below.
|Source||Number of Parameters (Millions)||Test Error||Search Method||Evaluation Method|
|PyramidNet + ShakeDrop||Yamada et al., 2018||26||2.31||Human designed||-|
|NASNet-A + cutout||Zoph et al., 2017||3.3||2.65||Reinforcement Learning||Full Train|
|AmoebaNet-B + cutout||Real et al., 2018||34.9||2.13||Evolutionary||Full Train|
|NAONET||Luo et al., 2018||28.6||2.98||Gradient||Partial Train|
|DARTS + cutout||H. Liu et al., 2018||3.4||2.83||Gradient||Weight Sharing|
For the CIFAR-10 benchmark, specialized NAS methods that use full training perform comparably to manually designed architectures, however they are prohibitively expensive and take over 1000 GPU days. Although methods that exploit partial training or other NAS-specific evaluation methods require less computation to perform the search (400 GPU days and ~1 GPU day, respectively), they are outperformed by the manually designed architecture in Table 1. Notably, the NAS architectures have nearly an order-of-magnitude fewer parameters than the human designed model, indicating promising applications of NAS to memory and latency constrained settings.
The test perplexity for different architectures on the PTB dataset are shown in Table 2.
|Source||Test Perplexity||Search Method||Evaluation Method|
|LSTM with MoS||Yang et al., 2017||54.4||Human designed||-|
|NASNet||Zoph et al., 2016||62.4||Reinforcement Learning||Full Train|
|NAONET||Luo et al., 2018||56.0||Gradient||Partial Train|
|DARTS||H. Liu et al., 2018||55.7||Gradient||Weight Sharing|
The specialized NAS results are less competitive on the PTB benchmark compared to manually designed architectures. It is surprising, however that cheaper evaluation methods outperform full training on this benchmark; this is likely due to the additional advances that have been made in training LSTMs since the publication of Zoph et.al., 2016.
Not yet! To be clear, exploring various architectures and performing extensive hyperparameter optimization remain crucial components of any deep learning application workflow. However, in light of the existing research results (as highlighted above), we believe that while specialized NAS methods have demonstrated promising results on these two benchmarks, they are still not ready for prime time for the following reasons:
Are you interested in accelerating deep learning model development by applying AutoML techniques to streamline your workflows, manage your GPU clusters, and track the work of your entire team in one place? Drop us an email at email@example.com to get started.
1 To learn more about ASHA, please see our latest blog post on Massively Parallel Hyperparameter Optimization.
2 For a detailed overview of NAS, we recommend the excellent survey by Elsken et al., 2017.
3 NAS focuses on the problem of identifying architectures, but nonetheless requires a secondary hyperparameter optimization step to tune the non-architecture-specific hyperparameters of the architecture it identifies. Our results show the test error after performing both steps.