Search

Majority voting and averaging are common approaches employed to resolve annotator disagreements and derive single ground truth labels from multiple annotations. However, annotators may systematically disagree with one another, often reflecting their individual biases and values, especially in the...

Standard machine learning is unable to accommodate inputs which do not belong to the training distribution. The resulting models often give rise to confident incorrect predictions which may lead to devastating consequences. This problem is especially demanding in the context of dense prediction...

Many subfields of machine learning share a common stumbling block: evaluation. Advances in machine learning often evaporate under closer scrutiny or turn out to be less widely applicable than originally hoped. We conduct a meta-review of 107 survey papers from natural language processing...

Background

Single-cell RNA sequencing (scRNA-seq) data provide valuable insights into cellular heterogeneity which is significantly improving the current knowledge on biology and human disease. One of the main applications of scRNA-seq data analysis is the identification of new cell types and...

MIT researchers are testing a simplified turbulence theory’s ability to model complex plasma phenomena using a novel machine-learning technique.

De novo protein design with deep learning can open new doors for medicine and many other fields

David Gamarnik has developed a new tool, the overlap gap property, for understanding computational problems that appear intractable.

In recent years, self-supervised learning has had significant success in applications involving computer vision and natural language processing. The type of pretext task is important to this boost in performance. One common pretext task is the measure of similarity and dissimilarity between pairs of...

Data science education is increasingly becoming an integral part of many educational structures, both informal and formal. Much of the attention has been on the application of AI principles and techniques, especially machine learning, natural language processing and predictive analytics. While AI is...

Over the past 60 years, artificial intelligence (AI) has made significant progress, but most of its benefits have failed to make a significant impact within the Global South. Current practices that have led to biased systems will prevent AI from being actualized unless significant efforts are made...

Researchers develop a way to test whether popular methods for understanding machine-learning models are working correctly.

Older adults are increasingly using technologies in their everyday lives, but the needs of this population are often ignored in AI design.

In this paper, we study the convergence properties of a randomized block-coordinate descent algorithm for the minimization of a composite convex objective function, where the block-coordinates are updated asynchronously and randomly according to an arbitrary probability distribution. We prove that...

In the recent CASP (Critical Assessment of Structure Prediction) competition, AlphaFold2 performed outstandingly. Its worst predictions were for NMR structures, which has two alternative explanations: either the NMR structures were poor, implying that AlphaFold may be more accurate than NMR; or...

The quest for determinism in machine learning has disproportionately focused on characterizing the impact of noise introduced by algorithmic design choices. In this work, we address a less well understood and studied question: how does our choice of tooling introduce randomness to deep neural...

Across fields of science, researchers have increasingly focused on designing soft devices that can shape-morph to achieve functionality. However, identifying a rest shape that leads to a target 3D shape upon actuation is a non-trivial task that involves inverse design capabilities. In this study, a...

Researchers have created a method to help workers collaborate with artificial intelligence systems.

ML is being deployed in complex, real-world scenarios where errors have impactful consequences. In these systems, thorough testing of the ML pipelines is critical. A key component in ML deployment pipelines is the curation of labeled training data. Common practice in the ML literature assumes that...

Even after decades of intensive research and public debates, the topic of data privacy remains surrounded by confusion and misinformation. Many people still struggle to grasp the importance of privacy, which has far-reaching consequences for social norms, jurisprudence, and legislation. Discussions...

Research in machine learning (ML) has argued that models trained on incomplete or biased datasets can lead to discriminatory outputs. In this commentary, we propose moving the research focus beyond bias-oriented framings by adopting a power-aware perspective to "study up" ML datasets. This means...