illustrated icons related to artificial intelligence for Microsoft's involvement at NeurIPS 2020
2020年12月6日 – 2020年12月12日

Microsoft at NeurIPS 2020

地点： Virtual

微软研究院博客

An visual comparing public image classification APIs to the proposed denoised smoothing framework applied to a public image classification API. In the first sequence, an image of an elephant is input into a public image classification API, represented by an arrow leading to a gray square labeled as such. An arrow from the square points to a correct prediction of elephant, enclosed in a green square, with the words “no robustness guaranteed” under it. In the second sequence, an arrow points from an image of an elephant to six noisy copies of the image. An arrow then points from the copies to a square labeled “Custom-trained Denoiser,” which outputs six clean versions of the images. An arrow points from the clean copies to a square labeled “Public Image Classification API.” The classifier provides predictions for each copy, of which four correctly identify their respective images as elephant. Adjacent to the predictions is a pie chart labeled “Majority Rules!” with one-third of the pie in red and two-thirds in green. Arrows point to the output of the process: a final prediction of elephant, enclosed in a green square, and a strong robustness guarantee, denoted by the words “certified radius” enclosed in a green square.
微软研究院博客

| Hadi Salman

Editor’s note: This post and its researc…

微软研究院博客

| Oindrila Saha, Shikhar Jaiswal, Aayan Kumar, Harsha Simhadri, 和 Prateek Jain

Computer vision has rapidly evolved over…

微软研究院博客

| Xu Tan

Pretrained language models have been a h…

diagram, schematic
微软研究院博客

As human beings, we encounter unfamiliar…

a man sitting at a table using a laptop
微软研究院博客

| Daniel McDuff 和 Xin Liu

According to the CDC WONDER Online Datab…

微软研究院博客

| Bita Darvish Rouhani, Doug Burger, Eric Chung, Rangan Majumder, Sangeetha Shekar, Saurabh Tiwary, Sitaram Lanka, 和 Steve Reinhardt

AI is taking on an increasingly importan…

Graphical representation of how our adversarial training approach works for solving the instrumental variable problem. The problem is viewed as a zero-sum game between a learner and an adversary. The learner attempts to find models that satisfy all moment constraints and the adversary flags violating moment constraints. Then the learner tries to correct the model to also satisfy the flagged constraint. A good model is learned when the adversary cannot find large violations
微软研究院博客

| Vasilis Syrgkanis

We are going through a new shift in mach…

微软研究院博客

| Hadi Salman

Editor’s note: This post and its researc…