تقرير
DBsurf: A Discrepancy Based Method for Discrete Stochastic Gradient Estimation
العنوان: | DBsurf: A Discrepancy Based Method for Discrete Stochastic Gradient Estimation |
---|---|
المؤلفون: | Arabi, Pau Mulet, Flowers, Alec, Mauch, Lukas, Cardinaux, Fabien |
سنة النشر: | 2023 |
المجموعة: | Computer Science |
مصطلحات موضوعية: | Computer Science - Machine Learning, I.2.0 |
الوصف: | Computing gradients of an expectation with respect to the distributional parameters of a discrete distribution is a problem arising in many fields of science and engineering. Typically, this problem is tackled using Reinforce, which frames the problem of gradient estimation as a Monte Carlo simulation. Unfortunately, the Reinforce estimator is especially sensitive to discrepancies between the true probability distribution and the drawn samples, a common issue in low sampling regimes that results in inaccurate gradient estimates. In this paper, we introduce DBsurf, a reinforce-based estimator for discrete distributions that uses a novel sampling procedure to reduce the discrepancy between the samples and the actual distribution. To assess the performance of our estimator, we subject it to a diverse set of tasks. Among existing estimators, DBsurf attains the lowest variance in a least squares problem commonly used in the literature for benchmarking. Furthermore, DBsurf achieves the best results for training variational auto-encoders (VAE) across different datasets and sampling setups. Finally, we apply DBsurf to build a simple and efficient Neural Architecture Search (NAS) algorithm with state-of-the-art performance. Comment: 22 pages, 7 figures |
نوع الوثيقة: | Working Paper |
URL الوصول: | http://arxiv.org/abs/2309.03974 |
رقم الأكسشن: | edsarx.2309.03974 |
قاعدة البيانات: | arXiv |
الوصف غير متاح. |