Change of variables: Apply $\tanh$ to the Gaussian samples
In the paper "Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor" Appendix C, it mentioned that applying $\tanh$ to the Gaussian sample gives us the
normal distribution - Change of variables: Apply $\tanh$ to the Gaussian samples - Mathematics Stack Exchange Stack Internal Knowledge at work Bring the best of human thought and AI automation together at your work. Explore Stack Internal Change of variables: Apply $\tanh$ to the Gaussian samples Ask Question Asked 7 years, 5 months ago Modified 7 years ago Viewed 3k times 6 $\begingroup$ In the paper "Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor" Appendix C, it mentioned that applying $\tanh$ to the Gaussian sample gives us the probability
related reading
- Some Math behind Neural Tangent Kernel | Lil'Loglilianweng.github.io
- Pen and Paper Exercises in Machine Learningarxiv.org
- Normalizing flowsbehindml.com
- Statistical Mechanics of Deep Learningganguli-gang.stanford.edu
- Correct definition of convolution of distributions? - Mathematics Stack Exchangemath.stackexchange.com
- An Introduction to Malliavin Calculusuni-ulm.de
- [2605.01172] A Theory of Generalization in Deep Learningarxiv.org
- fourier analysis - Convolution theorem with distributions - Mathematics Stack Exchangemath.stackexchange.com
- Understanding the Neural Tangent Kernel – EigenTaleseigentales.com
- Sub-Gaussian distribution - Wikipediaen.wikipedia.org
- How to Transform Data to Better Fit The Normal Distribution - MachineLearningMastery.commachinelearningmastery.com
- 2403.18103.pdfarxiv.org