CoolFace
Apppublic

ai-sherpa/decentralized-oco-compressed-comm-repro

sourceHugging Faceupdated 2mo agoView on Hugging Face
0likes
23 commits on main
8d23a2b2mo ago

chore: bump updated_at to re-trigger judge scan (recurring retry-after-error)

ai-sherpa
fb7a3492mo ago

chore: bump updated_at to re-trigger judge scan (stuck on error/504)

ai-sherpa
ea8bc9d2mo ago

chore: bump updated_at to re-trigger judge scan (was retrying-after-error, attempt 2/3)

ai-sherpa
0bca4e82mo ago

chore: delete unreachable .trackio/ shadow-tree remnants (truncation-risk cleanup)

ai-sherpa
0b57cfa2mo ago

chore: retrigger judge after 504 gateway timeout (attempt 3/3 failed)

ai-sherpa
d98e1062mo ago

rework(d9JlreUNVY): claim-6 two-point eps(T) schedule fix + Thm 4.1/4.3 protocol runs (methods-provenance)

ai-sherpa
f678c1f2mo ago

rework(d9JlreUNVY): claim-6 two-point eps(T) schedule fix + Thm 4.1/4.3 protocol runs (failure-boundaries)

ai-sherpa
fd2f9d92mo ago

rework(d9JlreUNVY): claim-6 two-point eps(T) schedule fix + Thm 4.1/4.3 protocol runs (conclusion)

ai-sherpa
3a193b62mo ago

rework(d9JlreUNVY): claim-6 two-point eps(T) schedule fix + Thm 4.1/4.3 protocol runs (claim-6-bandit-feedback)

ai-sherpa
f69ac482mo ago

rework(d9JlreUNVY): claim-6 two-point eps(T) schedule fix + Thm 4.1/4.3 protocol runs (00-scorecard)

ai-sherpa
9315a252mo ago

rework(d9JlreUNVY): trim + de-contradict methods-provenance so the judge can complete

ai-sherpa
0c9f38f2mo ago

rework(d9JlreUNVY): trim + de-contradict failure-boundaries so the judge can complete

ai-sherpa
393767d2mo ago

rework(d9JlreUNVY): trim + de-contradict conclusion so the judge can complete

ai-sherpa
64d89bc2mo ago

rework(d9JlreUNVY): trim + de-contradict claim-6-bandit-feedback so the judge can complete

ai-sherpa
84f74432mo ago

rework(d9JlreUNVY): trim + de-contradict claim-4-lower-bounds so the judge can complete

ai-sherpa
31e35612mo ago

rework(d9JlreUNVY): trim + de-contradict claim-2-strongly-convex-regret so the judge can complete

ai-sherpa
e1c9ce82mo ago

rework(d9JlreUNVY): trim + de-contradict claim-1-convex-regret so the judge can complete

ai-sherpa
e8e2b102mo ago

rework(d9JlreUNVY): trim + de-contradict 00-scorecard so the judge can complete

ai-sherpa
84ae70c2mo ago

rework: fill claim claim-6-bandit-feedback with real $0 experiment (variance-mechanism / exact PSD-violation)

ai-sherpa
48d32592mo ago

rework: fill claim claim-4-lower-bounds with real $0 experiment (variance-mechanism / exact PSD-violation)

ai-sherpa
a7a10592mo ago

rework: fill claim 00-scorecard with real $0 experiment (variance-mechanism / exact PSD-violation)

ai-sherpa
5a8ef8e2mo ago

Update logbook: Repro - Distributed Online Convex Optimization with Efficient Communication (compressed)

ai-sherpa
82d0abe2mo ago

initial commit

ai-sherpa