no code implementations • 4 May 2024 • Jingwei Zhang, Mohammad Jalali, Cheuk Ting Li, Farzan Farnia
An interpretable comparison of generative models requires the identification of sample types produced more frequently by each of the involved models.
no code implementations • 27 Feb 2024 • Jingwei Zhang, Cheuk Ting Li, Farzan Farnia
The massive developments of generative model frameworks and architectures require principled methods for the evaluation of a model's novelty compared to a reference dataset or baseline generative models.
no code implementations • 31 Oct 2023 • Mahmoud Hegazy, Rémi Leluc, Cheuk Ting Li, Aymeric Dieuleveut
Compression schemes have been extensively used in Federated Learning (FL) to reduce the communication cost of distributed learning.
1 code implementation • 29 Jan 2021 • Cheuk Ting Li
We present a versatile automated theorem proving framework capable of automated discovery, simplification and proofs of inner and outer bounds in network information theory, deduction of properties of information-theoretic quantities (e. g. Wyner and G\'acs-K\"orner common information), and discovery of non-Shannon-type inequalities, under a unified framework.
Automated Theorem Proving Information Theory Information Theory
1 code implementation • 13 Aug 2020 • Cheuk Ting Li
sequence of random variables $Z_{1},\ldots, Z_{n}$ that contains the same information as $X$, i. e., there exists an injective function $f$ such that $X=f(Z_{1},\ldots, Z_{n})$.
Information Theory Information Theory Probability 94A15, 60F05
1 code implementation • 14 Jun 2020 • Cheuk Ting Li
More precisely, we construct a coupling with entropy within 2 bits from the entropy of the greatest lower bound of $p_{1},\ldots, p_{m}$ with respect to majorization.
Information Theory Information Theory Probability
no code implementations • 21 Dec 2014 • Kim-Hung Li, Cheuk Ting Li
We learn from this phenomenon that when the size of the training data is large, we should either relax the assumption or apply NB to a "reduced" data set, say for example use NB as a local model.