$MUON muon

Latest papers

1. Measurement of electromagnetic transition form factors in two-photon collisions at BESIII hep-ex · 14 Oct · 1 hit

   The muon anomalous magnetic moment, $a_\mu = (g-2)_\mu/2$, shows a sizeable discrepancy between Standard Model…

2. Long-Horizon RL Needs Muon cs.LG · 14 Oct · 2 hits

   …policy while rollouts are still being collected. We examine a single-rollout asynchronous recipe combining Muon, token-averaged loss, truncated importance sampling, and auxiliary length and tool-error penalties.…

3. A low-cost compact muon detector for millikelvin operation physics.ins-det · 14 Oct · 8 hits

   …devices, such as kinetic inductance detectors and superconducting nanowire single-photon detectors. Cosmic muons are a component of this background that can only be partly shielded against at ground level, but can be…

4. From Spectral to Frobenius: When Muon Loses Its Spectral Implicit Bias cs.LG · 14 Oct · 5 hits

   …established spectral max-margin behavior for spectral-norm steepest descent methods, including idealized Muon with exact orthogonalization. In practice, however, Muon approximates orthogonalization using finite…

5. Memory-Efficient Subspace Optimization with Residual Correction: From Momentum SGD to Muon cs.LG · 14 Oct · 3 hits

   …correction approach that aligns them across stages. Beyond momentum SGD, we incorporate our framework into Muon and propose PMRC, which applies the matrix sign to the projected momentum. Through a weighted momentum…

6. How Inner Optimizers Shape Replica Scaling cs.LG · 13 Oct · 1 hit

   …decreasing performance. Existing work has empirically demonstrated that the choice of inner optimizer (Muon v.s. AdamW) can significantly improve this trend, but offers only limited explanation of the performance…

7. Which Optimizer Leaks Most? Optimizer Implicit Bias as a Training-Data Privacy Controller cs.LG · 13 Oct · 3 hits

   …to training-data reconstruction, we show that the implicit biases of popular optimizers (SGD, Adam, Muon) leave very different fingerprints of the training data. We argue that the key distinction is the…

8. NBCLT: Parameter-Agnostic Norm-Ball Clipped Local Training cs.LG · 13 Oct · 1 hit

   …tuned hyperparameters under heterogeneous data and communication constraints. At the same time, recent Muon-type optimizers suggest that constraining updates within suitable norm balls provides a simple way to…