We utilize neural network embeddings to detect data drift by formulating the drift detection within an appropriate sequential decision framework. This enables control of the false alarm rate although the statistical tests are repeatedly applied. Since change detection algorithms naturally face a tradeoff between avoiding false alarms and quick correct detection, we introduce a loss function which evaluates an algorithm's ability to balance these two concerns, and we use it in a series of experiments.

Objective. Supervised learning paradigms are often limited by the amount of labeled data that is available. This phenomenon is particularly problematic in clinically-relevant data, such as electroencephalography (EEG), where labeling can be costly in terms of specialized expertise and human processing time. Consequently, deep learning architectures designed to learn on EEG data have yielded relatively shallow models and performances at best similar to those of traditional feature-based approaches. However, in most situations, unlabeled data is available in abundance. By extracting information from this unlabeled data, it might be possible to reach competitive performance with deep neural networks despite limited access to labels. Approach. We investigated self-supervised learning (SSL), a promising technique for discovering structure in unlabeled data, to learn representations of EEG signals. Specifically, we explored two tasks based on temporal context prediction as well as contrastive predictive coding on two clinically-relevant problems: EEG-based sleep staging and pathology detection. We conducted experiments on two large public datasets with thousands of recordings and performed baseline comparisons with purely supervised and hand-engineered approaches. Main results. Linear classifiers trained on SSL-learned features consistently outperformed purely supervised deep neural networks in low-labeled data regimes while reaching competitive performance when all labels were available. Additionally, the embeddings learned with each method revealed clear latent structures related to physiological and clinical phenomena, such as age effects. Significance. We demonstrate the benefit of self-supervised learning approaches on EEG data. Our results suggest that SSL may pave the way to a wider use of deep learning models on EEG data.

In recent times, deep artificial neural networks have achieved many successes in pattern recognition. Part of this success is the reliance on big data to increase generalization. However, in the field of time series recognition, many datasets are often very small. One method of addressing this problem is through the use of data augmentation. In this paper, we survey data augmentation techniques for time series and their application to time series classification with neural networks. We outline four families of time series data augmentation, including transformation-based methods, pattern mixing, generative models, and decomposition methods, and detail their taxonomy. Furthermore, we empirically evaluate 12 time series data augmentation methods on 128 time series classification datasets with 6 different types of neural networks. Through the results, we are able to analyze the characteristics, advantages and disadvantages, and recommendations of each data augmentation method. This survey aims to help in the selection of time series data augmentation for neural network applications.

This paper proposes a new optimizer for deep learning, named d-AmsGrad. In the real-world data, noise and outliers cannot be excluded from dataset to be used for learning robot skills. This problem is especially striking for robots that learn by collecting data in real time, which cannot be sorted manually. Several noise-robust optimizers have therefore been developed to resolve this problem, and one of them, named AmsGrad, which is a variant of Adam optimizer, has a proof of its convergence. However, in practice, it does not improve learning performance in robotics scenarios. This reason is hypothesized that most of robot learning problems are non-stationary, but AmsGrad assumes the maximum second momentum during learning to be stationarily given. In order to adapt to the non-stationary problems, an improved version, which slowly decays the maximum second momentum, is proposed. The proposed optimizer has the same capability of reaching the global optimum as baselines, and its performance outperformed that of the baselines in robotics problems.

There is a longstanding debate whether the Kolmogorov-Arnold representation theorem can explain the use of more than one hidden layer in neural networks. The Kolmogorov-Arnold representation decomposes a multivariate function into an interior and an outer function and therefore has indeed a similar structure as a neural network with two hidden layers. But there are distinctive differences. One of the main obstacles is that the outer function depends on the represented function and can be wildly varying even if the represented function is smooth. We derive modifications of the Kolmogorov-Arnold representation that transfer smoothness properties of the represented function to the outer function and can be well approximated by ReLU networks. It appears that instead of two hidden layers, a more natural interpretation of the Kolmogorov-Arnold representation is that of a deep neural network where most of the layers are required to approximate the interior function.

Caching techniques are widely used in the era of cloud computing from applications, such as Web caches to infrastructures, Memcached and memory caches in computer architectures. Prediction of cached data can greatly help improve cache management and performance. The recent advancement of deep learning techniques enables the design of novel intelligent cache replacement policies. In this work, we propose a learning-aided approach to predict future data accesses. We find that a powerful LSTM-based recurrent neural network model can provide high prediction accuracy based on only a cache trace as input. The high accuracy results from a carefully crafted locality-driven feature design. Inspired by the high prediction accuracy, we propose a pseudo OPT policy and evaluate it upon 13 real-world storage workloads from Microsoft Research. Results demonstrate that the new cache policy improves state-of-art practical policies by up to 19.2% and incurs only 2.3% higher miss ratio than OPT on average.

Predictive modeling often uses black box machine learning methods, such as deep neural networks, to achieve state-of-the-art performance. In scientific domains, the scientist often wishes to discover which features are actually important for making the predictions. These discoveries may lead to costly follow-up experiments and as such it is important that the error rate on discoveries is not too high. Model-X knockoffs enable important features to be discovered with control of the FDR. However, knockoffs require rich generative models capable of accurately modeling the knockoff features while ensuring they obey the so-called "swap" property. We develop Deep Direct Likelihood Knockoffs (DDLK), which directly minimizes the KL divergence implied by the knockoff swap property. DDLK consists of two stages: it first maximizes the explicit likelihood of the features, then minimizes the KL divergence between the joint distribution of features and knockoffs and any swap between them. To ensure that the generated knockoffs are valid under any possible swap, DDLK uses the Gumbel-Softmax trick to optimize the knockoff generator under the worst-case swap. We find DDLK has higher power than baselines while controlling the false discovery rate on a variety of synthetic and real benchmarks including a task involving a large dataset from one of the epicenters of COVID-19.

Given oracle access to a Neural Network (NN), it is possible to extract its underlying model. We here introduce a protection by adding parasitic layers which keep the underlying NN's predictions mostly unchanged while complexifying the task of reverse-engineering. Our countermeasure relies on approximating a noisy identity mapping with a Convolutional NN. We explain why the introduction of new parasitic layers complexifies the attacks. We report experiments regarding the performance and the accuracy of the protected NN.

So, you've seen some amazing GPT-3 demos on Twitter (if not, where have you been?). This mega machine learning model, created by OpenAI, can write it's own op-eds, poems, articles, and even working code: With GPT-3, I built a layout generator where you just describe any layout you want, and it generates the JSX code for you. GPT3()… the spreadsheet function to rule them all. Impressed with how well it pattern matches from a few examples. The same function looked up state populations, peoples' twitter usernames and employers, and did some math.

If you want a glimpse of the future, check out how developers are using gpt-3. This natural language processor was trained on parameters ten times greater than its most sophisticated rival and can be used to answer questions and write astoundingly well. Creative professionals everywhere, from top coders to professional writers marvel at what gpt-3 can produce even now – in its relative infancy. Yesterday, New York Times tech columnist Farhad Manjoo wrote that the short glimpse the general public has taken of gpt-3 "is at once amazing, spooky, humbling, and more than a little terrifying. GPT-3 is capable of generating entirely original, coherent, and sometimes even factual prose. And not just prose -- it can write poetry, dialogue, memes, computer code, and who knows what else." Manjoo speculated on whether a similar but more advanced AI might replace him someday.