Methods › Reinforcement Learning › Policy Gradient Methods › SVPG
Stein Variational Policy Gradient
SVPG
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Stein Variational Policy Gradient, or SVPG, is a policy gradient based method in reinforcement learning that uses Stein Variational Gradient Descent to allow simultaneous exploitation and exploration of multiple policies. Unlike traditional policy optimization which attempts to learn a single policy, SVPG models a distribution of policy parameters, where samples from this distribution will represent strong policies. SVPG optimizes this distribution of policy parameters with (relative) entropy regularization. The (relative) entropy term explicitly encourages exploration in the parameter space while also optimizing the expected utility of polices drawn from this distribution. Stein variational gradient descent (SVGD) is then used to optimize this distribution. SVGD leverages efficient deterministic dynamics to transport a set of particles to approximate given target posterior distributions.
The update takes the form:
$$ $$
∇θᵢ = 1/n∑ⱼ₌₁ⁿ ∇_(θⱼ) (1/α J(θⱼ) + logq₀(θⱼ))k(θⱼ, θᵢ) + ∇_(θⱼ) k(θⱼ, θᵢ)
Note that here the magnitude of α adjusts the relative importance between the policy gradient and the prior term ∇_(θⱼ) (1/α J(θⱼ) + logq₀(θⱼ))k(θⱼ, θᵢ) and the repulsive term ∇_(θⱼ) k(θⱼ, θᵢ). The repulsive functional is used to diversify particles to enable parameter exploration. A suitable α provides a good trade-off between exploitation and exploration. If α is too large, the Stein gradient would only drive the particles to be consistent with the prior q₀. As α→0, this algorithm is reduced to running n copies of independent policy gradient algorithms, if {θᵢ} are initialized very differently. A careful annealing scheme of α allows efficient exploration in the beginning of training and later focuses on exploitation towards the end of training.
In Syntology name not resolved (unknown_name). The lookup matched no method, paper or venue by this name; that asserts nothing about whether code exists.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
Papers archive 2025-07-28
archive papers tagged: 2 · with a code link: 1 · where Syntology ran a sample: 0 (0 with a run with no instrument failure, 0 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran: none of the 2 tagged papers has a Syntology run on record
Page 1 of 1: papers 1 to 2 of 2, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
Biologically Plausible Variational Policy Gradient with Spiking Recurrent Winner-Take-All Networks 21 Oct 2022 · 1 repository · arXiv:2210.13225
-
Stein Variational Policy Gradient 7 Apr 2017 · 0 repositories · arXiv:1704.02399
Tasks archive 2025-07-28
8 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
| Task | Papers |
|---|---|
| Reinforcement Learning (RL) | 2 |
| All | 1 |
| Bayesian Inference | 1 |
| Continuous Control | 1 |
| Policy Gradient Methods | 1 |
| Reinforcement Learning | 1 |
| continuous-control | 1 |
| reinforcement-learning | 1 |
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections