Neural Information Processing Systems http://nips.cc/

Improving the accuracy of deep neural networks on out-of-distribution (OOD) data is critical to an acceptance of deep learning in real world applications. It has been observed that accuracies on in-distribution (ID) versus OOD data follow a linear trend and models that outperform this baseline are exceptionally rare (and referred to as ``effectively robust"). Recently, some promising approaches have been developed to improve OOD robustness: model pruning, data augmentation, and ensembling or zero-shot evaluating large pretrained models. However, there still is no clear understanding of the conditions on OOD data and model properties that are required to observe effective robustness. We approach this issue by conducting a comprehensive empirical study of diverse approaches that are known to impact OOD robustness on a broad range of natural and synthetic distribution shifts of CIFAR-10 and ImageNet. In particular, we view the effective robustness puzzle through a Fourier lens and ask how spectral properties of both models and OOD data correlate with OOD robustness. We find this Fourier lens offers some insight into why certain robust models, particularly those from the CLIP family, achieve OOD robustness. However, our analysis also makes clear that no known metric is consistently the best explanation of OOD robustness. Thus, to aid future research into the OOD puzzle, we address the gap in publicly-available models with effective robustness by introducing a set of pretrained CIFAR-10 models---$RobustNets$---with varying levels of OOD robustness.

Semantic Scene Completion (SSC) aims to perform geometric completion and semantic segmentation simultaneously. Despite the promising results achieved by existing studies, the inherently ill-posed nature of the task presents significant challenges in diverse driving scenarios. This paper introduces TALoS, a novel test-time adaptation approach for SSC that excavates the information available in driving environments. Specifically, we focus on that observations made at a certain moment can serve as Ground Truth (GT) for scene completion at another moment. Given the characteristics of the LiDAR sensor, an observation of an object at a certain location confirms both 1) the occupation of that location and 2) the absence of obstacles along the line of sight from the LiDAR to that point.

The rebuttal from the authors was concise. However, I was not convinced about the assumption in Eq. 14 of the paper and how the authors defended it. The authors say: "Many documents (text categorization),[..] or time series signals (speech recognition) in a training set are alike. This fact is not systematically exploited by any existing stochastic optimization method!" I don't think this is correct.

In my view, the main reason the proposed lower bound is interesting is that it offers a potential way to speed up training for multi-class models with a very large number of classes. While it is useful to understand other properties of the lower bound, the paper could be improved by emphasizing this primary use case in machine learning. Figure 1c and Figure 3 need a more clear explanation of what is being displayed, and why it is important. In particular, what value is being plotted on the y-axis, and at what setting of the parameters w. Here is how I understand it, for Figure 1c say: Blue Line - value of Eq. (13) at the setting of parameters w that maximize 13 Red Line - value of Eq. (13) at the setting of parameters w that maximize 14 Green Line - value of Eq. (13)? at the setting of parameters w that maximize the Bouchard lower bound (?) Red dashed line - value of Eq. (13)? at parameters w based on the given iterations of training?

Topics include Machine Learning, Deep Learning, GUI, Computer Vision, and API development. Author is a Machine Learning Researcher, currently working in IBM. Taught 20,000 students & working professionals to crack coding interviews and land their dream job.

When the new teaser trailer for Avatar: The Way of Water--the next entry in James Cameron's CGI-heavy film franchise--came out, many viewers opined that the footage resembles a video game. As praise or pejorative, that comparison is a touch hyperbolic. Yet it signals, too, the perceived overlap between the video game and film industries, which have increasingly come to share technological, narrative, and visual approaches. Multiplex screens are nowadays laden with game-like images--exceptions exist, but a sense of green-screened unreality certainly abounds, whether you're watching an explosion-rich action film or a well-paced drama. Other ideas also flow freely across mediums: Games and movies alike have set their watches to Matrix-style "bullet time" effects; both forms have shaken up their cameras à la Bourne; and as virtuosic a filmmaker as Brian De Palma has marveled at how certain games have deftly repurposed cinema's roaming, first-person point-of-view shots.

Are you an enthusiast in AI and are you searching for good examples of applications to practice? This article could interest you. There are a lot of different possible use cases for Artificial intelligence, and some of them are interesting. There are a lot of examples of the GPT-3 that creates images given a specific text, and some of them are fancy. This is to say that AI is really powerful nowadays, and it can perform a huge number of tasks. One of them is to generate a text, an article, or whatever given a prompt or a point where to start.

Now we can start writing the code. First, let's import packages and define the main training parameters: All images during the training processes will be resized to this size. Depending on your hardware, it might be necessary to use a smaller batchSize or image size to avoid out-of-memory problems. Note that since we train with only a single image size, the net once trained is likely to be limited to work with only images of this size. In most cases what you want to do is change the size of each training batch.

Clusters of conversation provide a more valuable way to spend ones time than attending sessions. At the last national meeting we escaped from the celebrations of the recent victory of Deep Blue over the dreaded Kasparov, to find just such a group, already engaged in an animated discussion: A: We need to draw a line. A: Between a program that has some intelligence in it and one that doesn't. All Deep Blue does is brute-force search. That hardly counts as AI.