[2002.05616] Cutting out the Middle-Man: Training and Evaluating Energy-Based Models without Samplingcontact arXivarXiv Twitter

We present a new method for evaluating and training unnormalized density models. Our approach only requires access to the gradient of the unnormalized model's log-density. We estimate the Stein discrepancy between the data density p(x) and the model density q(x) defined by a vector function of the data. We parameterize this function with a neural network and fit its parameters to maximize the discrepancy. This yields a novel goodness-of-fit test which outperforms existing methods on high dimensional data. Furthermore, optimizing $q(x)$ to minimize this discrepancy produces a novel method for training unnormalized models which scales more gracefully than existing methods. The ability to both learn and compare models is a unique feature of the proposed method.

3 mentions: @wgrathwohl@Montreal_AI
Date: 2020/02/14 03:51

Referring Tweets

@wgrathwohl HOMIEZ, check my new paper: t.co/xZjYtgCd31 we develop a new method for training and evaluating unnormalized density models called LSD, the Learned Stein Discrepancy. Joint work with @kcjacksonwang @jh_jacobsen @DavidDuvenaud and Richard Zemel t.co/SQMElvcXUb
@Montreal_AI Cutting out the Middle-Man: Training and Evaluating Energy-Based Models without Sampling Grathwohl et al.: t.co/8D4J5tmXwf #ArtificialIntelligence #DeepLearning #MachineLearning t.co/mwo9YB5Gi2