Recommendations as Treatments

Recommendations as Treatments: Debiasing Learning and Evaluation

Most data for evaluating and training recommender systems is subject to selection biases, either through self-selection by the users or through the actions of the recommendation system itself. In this paper, we provide a principled approach to handling selection biases, adapting models and estimation techniques from causal inference. The approach leads to unbiased performance estimators despite...

Repurposing ARBs as treatments for breast cancer

Dibenzoxepines as treatments for neurodegenerative diseases*

In recent years, apoptotic cell death has been implicated with different progressive neurodegenerative diseases such as Parkinson's disease, Huntington's disease, Amyotrophic Lateral Sclerosis or Alzheimer's disease. The hypothesis emerged, that a drug preventing apoptosis may slow or even halt the disease progression. (±)-Deprenyl was reported to rescue neurons from cell death in different in ...

FDA approved drugs as potential Ebola treatments

In the search for treatments for the Ebola Virus, multiple screens of FDA drugs have led to the identification of several with promising in vitro activity. These compounds were not originally developed as antivirals and some have been further tested in mouse in vivo models. We put forward the opinion that some of these drugs could be evaluated further and move into the clinic as they are alread...

CryptoRec: Secure Recommendations as a Service

Recommender systems rely on large datasets of historical data and entail serious privacy risks. A server offering recommendations as a service to a client might leak more information than necessary regarding its recommendation model and training dataset. At the same time, the disclosure of the client’s preferences to the server is also a matter of concern. Providing recommendations while preser...