The blind application of machine learning runs the risk of amplifying biases present in data.
Such a danger is facing us with word embedding, a popular framework to represent text data …

We present modality gap, an intriguing geometric phenomenon of the representation space
of multi-modal models. Specifically, we show that different data modalities (eg images and …

Offline reinforcement learning (RL) refers to the problem of learning policies entirely from a
batch of previously collected data. This problem setting is compelling, because it offers the …

Differences in chromatin organization are key to the multiplicity of cell states that arise from
a single genetic background, yet the landscapes of in vivo tissues remain largely uncharted. …

Interpretability has become an important topic of research as more machine learning (ML)
models are deployed and widely used to make important decisions. Most of the current …

Accurate assessment of cardiac function is crucial for the diagnosis of cardiovascular
disease 1 , screening for cardiotoxicity 2 and decisions regarding the clinical management of …

Mammalian gene regulation is dependent on tissue-specific enhancers that can act across
large distances to influence transcriptional activity 1 , 2 , 3 . Mapping experiments have …

Artificial intelligence (AI) has been developed for echocardiography 1 , 2 – 3 , although it
has not yet been tested with blinding and randomization. Here we designed a blinded, …

The ability to computationally generate novel yet physically foldable protein structures could
lead to new biological discoveries and new treatments targeting yet incurable diseases. …

Intense recent discussions have focused on how to provide individuals with control over
when their data can and cannot be used---the EU’s Right To Be Forgotten regulation is an …