Researchers at Rensselaer Polytechnic Institute and IBM Research in New York have recently created pFoodReQ, a system that can recommend recipes tailored around the preferences and dietary needs of individual users. This system is outlined in a paper pre-published on arXiv and set to be presented at the 14th International Conference on Web Search and Data Mining (WSDM) in March. "Our work focuses on personalized food recommendation," Mohammed J. Zaki, one of the researchers who developed the system, told TechXplore. "In particular, given a user query in natural language, we want to retrieve the top matches in a recipe dataset." The short-term goal of the study carried out by Zaki and his colleagues was to help people find healthy recipes that satisfy both their dietary needs and inclinations.

Interpretability in machine learning (ML) is crucial for high stakes decisions and troubleshooting. In this work, we provide fundamental principles for interpretable ML, and dispel common misunderstandings that dilute the importance of this crucial topic. We also identify 10 technical challenge areas in interpretable machine learning and provide history and background on each problem. Some of these problems are classically important, and some are recent problems that have arisen in the last few years. These problems are: (1) Optimizing sparse logical models such as decision trees; (2) Optimization of scoring systems; (3) Placing constraints into generalized additive models to encourage sparsity and better interpretability; (4) Modern case-based reasoning, including neural networks and matching for causal inference; (5) Complete supervised disentanglement of neural networks; (6) Complete or even partial unsupervised disentanglement of neural networks; (7) Dimensionality reduction for data visualization; (8) Machine learning models that can incorporate physics and other generative or causal constraints; (9) Characterization of the "Rashomon set" of good models; and (10) Interpretable reinforcement learning. This survey is suitable as a starting point for statisticians and computer scientists interested in working in interpretable machine learning.

Subsampling unconditional generative adversarial networks (GANs) to improve the overall image quality has been studied recently. However, these methods often require high training costs (e.g., storage space, parameter tuning) and may be inefficient or even inapplicable for subsampling conditional GANs, such as class-conditional GANs and continuous conditional GANs (CcGANs), when the condition has many distinct values. In this paper, we propose an efficient method called conditional density ratio estimation in feature space with conditional Softplus loss (cDRE-F-cSP). With cDRE-F-cSP, we estimate an image's conditional density ratio based on a novel conditional Softplus (cSP) loss in the feature space learned by a specially designed ResNet-34 or sparse autoencoder. We then derive the error bound of a conditional density ratio model trained with the proposed cSP loss. Finally, we propose a rejection sampling scheme, termed cDRE-F-cSP+RS, which can subsample both class-conditional GANs and CcGANs efficiently. An extra filtering scheme is also developed for CcGANs to increase the label consistency. Experiments on CIFAR-10 and Tiny-ImageNet datasets show that cDRE-F-cSP+RS can substantially improve the Intra-FID and FID scores of BigGAN. Experiments on RC-49 and UTKFace datasets demonstrate that cDRE-F-cSP+RS also improves Intra-FID, Diversity, and Label Score of CcGANs. Moreover, to show the high efficiency of cDRE-F-cSP+RS, we compare it with the state-of-the-art unconditional subsampling method (i.e., DRE-F-SP+RS). With comparable or even better performance, cDRE-F-cSP+RS only requires about \textbf{10}\% and \textbf{1.7}\% of the training costs spent respectively on CIFAR-10 and UTKFace by DRE-F-SP+RS.

To efficiently solve distributed online learning problems with complicated constraints, previous studies have proposed several distributed projection-free algorithms. The state-of-the-art one achieves the $O({T}^{3/4})$ regret bound with $O(\sqrt{T})$ communication complexity. In this paper, we further exploit the strong convexity of loss functions to improve the regret bound and communication complexity. Specifically, we first propose a distributed projection-free algorithm for strongly convex loss functions, which enjoys a better regret bound of $O(T^{2/3}\log T)$ with smaller communication complexity of $O(T^{1/3})$. Furthermore, we demonstrate that the regret of distributed online algorithms with $C$ communication rounds has a lower bound of $\Omega(T/C)$, even when the loss functions are strongly convex. This lower bound implies that the $O(T^{1/3})$ communication complexity of our algorithm is nearly optimal for obtaining the $O(T^{2/3}\log T)$ regret bound up to polylogarithmic factors. Finally, we extend our algorithm into the bandit setting and obtain similar theoretical guarantees.

We consider the vector embedding problem. We are given a finite set of items, with the goal of assigning a representative vector to each one, possibly under some constraints (such as the collection of vectors being standardized, i.e., have zero mean and unit covariance). We are given data indicating that some pairs of items are similar, and optionally, some other pairs are dissimilar. For pairs of similar items, we want the corresponding vectors to be near each other, and for dissimilar pairs, we want the corresponding vectors to not be near each other, measured in Euclidean distance. We formalize this by introducing distortion functions, defined for some pairs of the items. Our goal is to choose an embedding that minimizes the total distortion, subject to the constraints. We call this the minimum-distortion embedding (MDE) problem. The MDE framework is simple but general. It includes a wide variety of embedding methods, such as spectral embedding, principal component analysis, multidimensional scaling, dimensionality reduction methods (like Isomap and UMAP), force-directed layout, and others. It also includes new embeddings, and provides principled ways of validating historical and new embeddings alike. We develop a projected quasi-Newton method that approximately solves MDE problems and scales to large data sets. We implement this method in PyMDE, an open-source Python package. In PyMDE, users can select from a library of distortion functions and constraints or specify custom ones, making it easy to rapidly experiment with different embeddings. Our software scales to data sets with millions of items and tens of millions of distortion functions. To demonstrate our method, we compute embeddings for several real-world data sets, including images, an academic co-author network, US county demographic data, and single-cell mRNA transcriptomes.

The emergence of Graph Convolutional Network (GCN) has greatly boosted the progress of graph learning. However, two disturbing factors, noise and redundancy in graph data, and lack of interpretation for prediction results, impede further development of GCN. One solution is to recognize a predictive yet compressed subgraph to get rid of the noise and redundancy and obtain the interpretable part of the graph. This setting of subgraph is similar to the information bottleneck (IB) principle, which is less studied on graph-structured data and GCN. Inspired by the IB principle, we propose a novel subgraph information bottleneck (SIB) framework to recognize such subgraphs, named IB-subgraph. However, the intractability of mutual information and the discrete nature of graph data makes the objective of SIB notoriously hard to optimize. To this end, we introduce a bilevel optimization scheme coupled with a mutual information estimator for irregular graphs. Moreover, we propose a continuous relaxation for subgraph selection with a connectivity loss for stabilization. We further theoretically prove the error bound of our estimation scheme for mutual information and the noise-invariant nature of IB-subgraph. Extensive experiments on graph learning and large-scale point cloud tasks demonstrate the superior property of IB-subgraph.

You can also install and use Python from the command line, and you can use an alternative Python development environment such as Visual Studio Code with the Python Extension. The version of Python on the client computer must match the version of Python on the server, and packages you install must be compliant with the version of Python you have. For information on which version of Python is included with each SQL Server version, see Python and R versions. To verify the version of Python on a particular SQL Server instance, use the following T-SQL command.

Deep learning is a powerful new technology, and we believe it should be applied across many disciplines. Domain experts are the most likely to find new applications of it, and we need more people from all backgrounds to get involved and start using it. That's why Jeremy cofounded fast.ai, to make deep learning easier to use through free online courses and software. Sylvain is a research engineer at Hugging Face. Previously he was a research scientist at fast.ai and a former mathematics and computer science teacher in a program that prepares students for entry into France's elite universities.

AI Researcher, Cognitive Technologist Inventor - AI Thinking, Think Chain Innovator - AIOT, XAI, Autonomous Cars, IIOT Founder Fisheyebox Spatial Computing Savant, Transformative Leader, Industry X.0 Practitioner A better way to build #machinelearning - why you should be using ActiveLearning? You spend less time and money on labelling data: Active learning has been shown to deliver large savings in data labelling across a wide range of tasks and data sets ranging from #computervision to #NLP. Since data labelling is one of the most expensive parts of training modern machine learning models this should be enough justification on its own. You get faster feedback on model performance: Usually people label their data before they train any models or get any feedback. Often, it takes days or weeks of iterating on #annotation guidelines and re-labelling only to discover that model performance falls far short of what is needed, or different labelled data is required.

GMRES is a powerful numerical solver used to find solutions to extremely large systems of linear equations. These systems of equations appear in many applications in science and engineering. Here we demonstrate a real-time machine learning algorithm that can be used to accelerate the time-to-solution for GMRES. Our framework is novel in that is integrates the deep learning algorithm in an in situ fashion: the AI-accelerator gradually learns how to optimizes the time to solution without requiring user input (such as a pre-trained data set). We describe how our algorithm collects data and optimizes GMRES. We demonstrate our algorithm by implementing an accelerated (MLGMRES) solver in Python. We then use MLGMRES to accelerate a solver for the Poisson equation -- a class of linear problems that appears in may applications. Informed by the properties of formal solutions to the Poisson equation, we test the performance of different neural networks. Our key takeaway is that networks which are capable of learning non-local relationships perform well, without needing to be scaled with the input problem size, making them good candidates for the extremely large problems encountered in high-performance computing. For the inputs studied, our method provides a roughly 2$\times$ acceleration.