pith:JXL4H4BW
A PAC-Bayes Approach for Controlling Unknown Linear Discrete-time Systems
A PAC-Bayes bound gives high-probability performance guarantees for any stochastic controller learned on unknown linear discrete-time systems.
arxiv:2605.10493 v2 · 2026-05-11 · math.OC · cs.SY · eess.SY · stat.ML
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{JXL4H4BW6PHSIROKSGWUO2333W}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
We derive a data-dependent high probability bound on the performance of any learned (stochastic) controller, and propose novel efficient learning algorithms with theoretical guarantees, which can be implemented for both finite and infinite controller spaces. Compared to prior work, our bound holds for unbounded quadratic cost.
The system parameters are drawn from a fixed but unknown distribution, and the controller is allowed to be stochastic; if the true parameter distribution changes over time or if only deterministic controllers are permitted, the derived bound and algorithms no longer apply.
A PAC-Bayes method supplies high-probability bounds on the cost of any learned stochastic controller for unknown linear systems and gives efficient algorithms that work for both finite and infinite controller sets, including unbounded quadratic costs.
Formal links
Receipt and verification
| First computed | 2026-05-22T01:03:20.127932Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
4dd7c3f036f3cf2445ca91ad476b7bddae2eb9923a71f4bdcc8c73f5488e23d6
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/JXL4H4BW6PHSIROKSGWUO2333W \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 4dd7c3f036f3cf2445ca91ad476b7bddae2eb9923a71f4bdcc8c73f5488e23d6
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "deaf1ca73ddb14fba7ec8c4a1c5333ec960dac5fedcd8cb46cb2f4bce9d74935",
"cross_cats_sorted": [
"cs.SY",
"eess.SY",
"stat.ML"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "math.OC",
"submitted_at": "2026-05-11T12:51:13Z",
"title_canon_sha256": "6d72184e0ab33cde3ca6d29048eb457709801f43b14419d2ef63f3fabcca642d"
},
"schema_version": "1.0",
"source": {
"id": "2605.10493",
"kind": "arxiv",
"version": 2
}
}