This work proposes a Gram-based anisotropic objective perturbation, a ``pre-distortion"strategy that counteracts the distortion from the covariate structure to restore isotropy in the estimation process and provides theoretical insights into designing stable and efficient private estimators without relying on data-depe...
This approach provides a unified framework to handle MLP within both classical and quantum LLMs and shows that the Pythagorean inequality continues to hold in the infinite-dimensional setting whenever the convex information-projection problem attains a finite minimum.
S. Sreekumar, Nir Weinberger· arXiv.org· 0 citations
This work proposes a modular and provably accurate cut-Bayes approach requiring no access to upstream data or model, and establishes, to the knowledge, first results on uniform Kullback-Leibler approximation rates of conditional distributions by common flow classes, yielding widely applicable fixed-data error rates for...
DLM-One eliminates iterative refinement by aligning the scores of a student model's outputs with the score function of a pretrained teacher DLM in the forward-diffused noisy space, which positions one-step score distillation as a viable path for the efficient deployment of continuous diffusion models operating in conti...
Tianqi Chen, Shujian Zhang, Mi Zhou· arXiv.org· 16 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Matrix-variate distributions are a relatively recent addition to the model-based clustering literature, thereby making it possible to analyze data in matrix form with complex structure such as images and time series. Due to its recent appearance, there is limited literature on matrix-variate data, with even less on dea...
Katharine M. Clark, Paul D. McNicholas· 0 citations
Thompson sampling (TS) is one of the most popular and earliest algorithms to solve stochastic multi-armed bandit problems. We consider a variant of TS, named $\alpha$-TS, where we use a fractional or $\alpha$-posterior ($\alpha\in(0,1)$) instead of the standard posterior distribution. To compute an $\alpha$-posterior,...
Prateek Jaiswal, Debdeep Pati, Anirban Bhattacharya et al.· 0 citations
This work analyzes the convergence of the noisy power method and Oja's algorithm, both studied for the stationary data generating model, and argues that the noisy power method is rate-optimal in this setting.
D. Bienstock, Minchan Jeong, A. Shukla et al.· Neural Information Processin...· 8 citations· ⚡1
Canon, a risk-averse method designed to optimize Value-at-Risk (VaR) constrained reinforcement learning (RL) problems, is introduced and Cantelli's inequality is employed to obtain a tractable, conservative and smooth bound on the VaR constraint based on the first two moments of the cost return.
A novel sampling algorithm is presented that uses observed label frequencies to estimate multivariate Bernoulli distribution parameters and calculates weights for each label combination, to preserve category frequency order, reduce frequency differences between most and least common categories, and account for category...
Simon Chung, Colby J. Vorland, D. L. Maney et al.· arXiv.org· 2 citations
On a 124M-parameter language model and selected short training windows, the method can lower simulated ledger cost relative to the tested baselines when fleet work is sufficiently cheap, and can lower simulated ledger cost relative to the tested baselines when fleet work is sufficiently cheap.
K. Ciosek, Nicolò Felicioni, Juan Elenter et al.· 2 citations
The anytime lower bound shows that the rate of non-anytime silver schedules (Altschuler and Parrilo, 2025; Grimmer et al., 2025) is unattainable in the anytime setting, and establishes a strict separation between the two settings.
It is proved that in noiseless Gaussian compressed sensing, the full-dimensional linear prior attains the minimum expected reconstruction error over the entire family of linear priors, indicating that the experimental benefits of tunability in compressed sensing with neural network priors arises due to nonlinearities i...
A machine-learning approach called CheMeleon provides a practical way to develop new drugs or other molecules with much less detailed, expensive experimentation.
