Alert button
Picture for Gabriel Goh

Gabriel Goh

Alert button

Zero-Shot Text-to-Image Generation

Feb 26, 2021
Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Chelsea Voss, Alec Radford, Mark Chen, Ilya Sutskever

Figure 1 for Zero-Shot Text-to-Image Generation
Figure 2 for Zero-Shot Text-to-Image Generation
Figure 3 for Zero-Shot Text-to-Image Generation
Figure 4 for Zero-Shot Text-to-Image Generation

Text-to-image generation has traditionally focused on finding better modeling assumptions for training on a fixed dataset. These assumptions might involve complex architectures, auxiliary losses, or side information such as object part labels or segmentation masks supplied during training. We describe a simple approach for this task based on a transformer that autoregressively models the text and image tokens as a single stream of data. With sufficient data and scale, our approach is competitive with previous domain-specific models when evaluated in a zero-shot fashion.

Viaarxiv icon

Learning Transferable Visual Models From Natural Language Supervision

Feb 26, 2021
Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever

Figure 1 for Learning Transferable Visual Models From Natural Language Supervision
Figure 2 for Learning Transferable Visual Models From Natural Language Supervision
Figure 3 for Learning Transferable Visual Models From Natural Language Supervision
Figure 4 for Learning Transferable Visual Models From Natural Language Supervision

State-of-the-art computer vision systems are trained to predict a fixed set of predetermined object categories. This restricted form of supervision limits their generality and usability since additional labeled data is needed to specify any other visual concept. Learning directly from raw text about images is a promising alternative which leverages a much broader source of supervision. We demonstrate that the simple pre-training task of predicting which caption goes with which image is an efficient and scalable way to learn SOTA image representations from scratch on a dataset of 400 million (image, text) pairs collected from the internet. After pre-training, natural language is used to reference learned visual concepts (or describe new ones) enabling zero-shot transfer of the model to downstream tasks. We study the performance of this approach by benchmarking on over 30 different existing computer vision datasets, spanning tasks such as OCR, action recognition in videos, geo-localization, and many types of fine-grained object classification. The model transfers non-trivially to most tasks and is often competitive with a fully supervised baseline without the need for any dataset specific training. For instance, we match the accuracy of the original ResNet-50 on ImageNet zero-shot without needing to use any of the 1.28 million training examples it was trained on. We release our code and pre-trained model weights at https://github.com/OpenAI/CLIP.

Viaarxiv icon

Satisfying Real-world Goals with Dataset Constraints

May 03, 2017
Gabriel Goh, Andrew Cotter, Maya Gupta, Michael Friedlander

Figure 1 for Satisfying Real-world Goals with Dataset Constraints
Figure 2 for Satisfying Real-world Goals with Dataset Constraints
Figure 3 for Satisfying Real-world Goals with Dataset Constraints
Figure 4 for Satisfying Real-world Goals with Dataset Constraints

The goal of minimizing misclassification error on a training set is often just one of several real-world goals that might be defined on different datasets. For example, one may require a classifier to also make positive predictions at some specified rate for some subpopulation (fairness), or to achieve a specified empirical recall. Other real-world goals include reducing churn with respect to a previously deployed model, or stabilizing online training. In this paper we propose handling multiple goals on multiple datasets by training with dataset constraints, using the ramp penalty to accurately quantify costs, and present an efficient algorithm to approximately optimize the resulting non-convex constrained optimization problem. Experiments on both benchmark and real-world industry datasets demonstrate the effectiveness of our approach.

Viaarxiv icon