Jordan Bryan

SoundCloud

Announcing the creation of a new SoundCloud profile. Tracks will be posted sporadically, with the hope of improvement in quality as time goes on. Instruments: Bass Trumpet Keyboard + Ableton libraries, packs and plugins Equipment: Ableton Live 10 Focusite Scarlett Solo Neewer NW-700 microphone Akai APC Key 25 Idols: Pu... Read more

Estimating signal and noise in linear models

Let \(\mathbf{Y} \in \mathbb{R}^n\) and \(\boldsymbol{\beta} \in \mathbb{R}^p\) and consider the model \[\begin{aligned} \mathbf{Y} | \boldsymbol{\beta} &\sim N(\mathbf{X} \boldsymbol{\beta}, \tau^2 \mathbf{I}) \\ \boldsymbol{\beta} &\sim N(0, \psi^2 \mathbf{I}) \end{aligned}\] The parameters \(\psi^2\) and \(\tau^2\) represent... Read more

Discovering the anticancer potential of non-oncology drugs by systematic viability profiling

New work from a group of researchers at the Broad Institute of MIT and Harvard unveils the cancer-fighting potential of non-cancer drugs. The article describes how the PRISM profiling method was used to to create viability profiles for over \(4000\) drugs across \(578\) human cancer cell lines. Many non-oncology drugs, including one previously a... Read more

Music in the wild

Neural networks have become popular tools for transfering the style of one artistic work or photograph to another. Methods for performing style transfer are distinguished from one another by features of the neural network’s design as well as the loss function that guides network updates. In a recent project, I teamed up with Alessandro Zito to u... Read more

Implementing minimum-Wasserstein deconvolution

Recently, Rigollet and Weed (2019)1 introduced a consistent estimator for the problem of uncoupled isotonic regression. The authors also proposed a computationally efficient procedure for computing a relaxed version of the estimator, which achieves the same minimax rate. Rigollet and Weed’s estimator relaxed estimator minimizes the Wasserstein-... Read more