3 problems

Consider quantizer designs in which local observations are mapped to quantizer outputs, with stationary or blockwise stationary designs allowed. A quantizer is likelihood-based whe…

Let ${\mathcal X}={\mathcal Y}={\mathbb R}$, let ${\mathsf P}_{X,Y}$ be a fixed distribution independent of $n$, and define the likelihood ratio … Let ${\mathcal L}$ be the likelih…

Let $\mathsf{H}_0,\mathsf{H}_1,\dots,\mathsf{H}_N$ be the hypotheses in a general non-i.i.d. stochastic model, and let $\lambda_{ij}(n)$ denote the log-likelihood ratio between…