Data interpreter: An llm agent for data science
Large Language Model (LLM)-based agents have shown effectiveness across many
applications. However, their use in data science scenarios requiring solving long-term …
applications. However, their use in data science scenarios requiring solving long-term …
Nas-bench-suite-zero: Accelerating research on zero cost proxies
Zero-cost proxies (ZC proxies) are a recent architecture performance prediction technique
aiming to significantly speed up algorithms for neural architecture search (NAS). Recent …
aiming to significantly speed up algorithms for neural architecture search (NAS). Recent …
[PDF][PDF] Data science at the singularity
Something fundamental to computation-based research has really changed in the last ten
years. In certain fields, progress is simply dramatically more rapid than previously …
years. In certain fields, progress is simply dramatically more rapid than previously …
Meta-album: Multi-domain meta-dataset for few-shot image classification
Abstract We introduce Meta-Album, an image classification meta-dataset designed to
facilitate few-shot learning, transfer learning, meta-learning, among other tasks. It includes …
facilitate few-shot learning, transfer learning, meta-learning, among other tasks. It includes …
Dual-view molecular pre-training
Molecular pre-training, which is about to learn an effective representation for molecules on
large amount of data, has attracted substantial attention in cheminformatics and …
large amount of data, has attracted substantial attention in cheminformatics and …
Human behavior in image-based Road Health Inspection Systems despite the emerging AutoML
Introduction The emergence of automated machine learning or AutoML has raised an
interesting trend of no-code and low-code machine learning where most tasks in the …
interesting trend of no-code and low-code machine learning where most tasks in the …
Lessons learned from the NeurIPS 2021 MetaDL challenge: Backbone fine-tuning without episodic meta-learning dominates for few-shot learning image classification
Although deep neural networks are capable of achieving performance superior to humans
on various tasks, they are notorious for requiring large amounts of data and computing …
on various tasks, they are notorious for requiring large amounts of data and computing …
Towards efficient AutoML: a pipeline synthesis approach leveraging pre-trained transformers for multimodal data
This paper introduces an Automated Machine Learning (AutoML) framework specifically
designed to efficiently synthesize end-to-end multimodal machine learning pipelines …
designed to efficiently synthesize end-to-end multimodal machine learning pipelines …
NeurIPS'22 cross-domain MetaDL challenge: Results and lessons learned
Deep neural networks have demonstrated the ability to outperform humans in multiple tasks,
but they often require substantial amounts of data and computational resources. These …
but they often require substantial amounts of data and computational resources. These …
Visda-2021 competition: Universal domain adaptation to improve performance on out-of-distribution data
Progress in machine learning is typically measured by training and testing a model on
samples drawn from the same distribution, ie the same domain. This over-estimates future …
samples drawn from the same distribution, ie the same domain. This over-estimates future …