Resultado da Busca
Há 3 dias · Right: Topics highlighted in the invited talk including prompt-based ML development (Andrew Ng), the DMLR ecosystem (Peter Mattson), reality-centric AI (Mihaela van der Schaar), bias in vision data (Olga Russakovsky and Vikram Ramaswamy), history of distribution shifts dating back to NeurIPS 2006 (Masashi Sugiyama), the AI research agent (Isabelle Guyon), nuances of data quality (Dina Machuve ...
Há 5 dias · Olga Russakovsky, Jia Deng, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, Andrej Karpathy, Aditya Khosla, Michael Bernstein, et al. Imagenet large scale visual recognition challenge. International journal of computer vision, 115(3):211-252, 2015. Google Scholar Digital Library
Há 1 dia · Olga Russakovsky; Andrés Monroy-Hernández; View. Building trust in automatic video interviews using various AI interfaces: Tangibility, immediacy, and transparency. Article. Feb 2023;
Há 5 dias · The analysis shows that these AI objects routinely traffic in normative gender roles of the feminine as caretaker, mother, and wife in order to obfuscate modes of surveillance, and mediate the relationship users and potential users have with late-capitalist market logics in the platform economy.
FineWeb-Edu, a subset of FineWeb constructed using scalable automated high-quality annotations for educational value, and which outperforms all openly accessible web-datasets on a number of educational benchmarks such as MMLU, ARC, and OpenBookQA. 📚 FineWeb-Edu is available in two sizes/filtering-level: 1.3 trillion (very high educational ...
Há 5 dias · 2.2 CNNs produce responses that are qualitatively similar to the MEG evoked components. As a model of the computations underlying the brain activity observed during the meg experiment, we used a vgg-11 ( Szegedy et al., 2015) network architecture, pretrained on ImageNet ( Russakovsky et al., 2015 ).
Há 5 dias · In this work, we propose a texture-guided feature compression method. It can simultaneously provide natural images and deep features for humans and machines, respectively, and fully utilizes texture images to remove feature redundancy. Specifically, we construct a two-layer encoding framework consisting of a feature layer and a texture layer.
