Offline optimization has recently emerged as an increasingly popular approach to mitigate the prohibitively expensive cost of online experimentation.

Offline optimization has recently emerged as an increasingly popular approach to mitigate the prohibitively expensive cost of online experimentation. The key idea is to learn a surrogate of the black-box function that underlines the target experiment using a static (offline) dataset of its previous input-output queries. Such an approach is, however, fraught with an out-of-distribution issue where the learned surrogate becomes inaccurate outside the offline data regimes. To mitigate this, existing offline optimizers have proposed numerous conditioning techniques to prevent the learned surrogate from being too erratic. Nonetheless, such conditioning strategies are often specific to particular surrogate or search models, which might not generalize to a different model choice. This motivates us to develop a model-agnostic approach instead, which incorporates a notion of model sharpness into the training loss of the surrogate as a regularizer. Our approach is supported by a new theoretical analysis demonstrating that reducing surrogate sharpness on the offline dataset provably reduces its generalized sharpness on unseen data. Our analysis extends existing theories from bounding generalized prediction loss (on unseen data) with loss sharpness to bounding the worst-case generalized surrogate sharpness with its empirical estimate on training data, providing a new perspective on sharpness regularization. Our extensive experimentation on a diverse range of optimization tasks also shows that reducing surrogate sharpness often leads to significant improvement, marking (up to) a noticeable 9.6% performance boost. Our code is publicly available at https://github.com/cuong-dm/IGNITE

Electronic Health Records present a valuable modality for driving personalized medicine, where treatment is tailored to fit individual-level differences. For this purpose, many data-driven machine learning and statistical models rely on the wealth of longitudinal EHRs to study patients' physiological and treatment effects. However, longitudinal EHRs tend to be sparse and highly missing, where missingness could also be informative and reflect the underlying patient's health status. Therefore, the success of data-driven models for personalized medicine highly depends on how the EHR data is represented from physiological data, treatments, and the missing values in the data. To this end, we propose a novel deep-learning model that learns the underlying patient dynamics over time across multivariate data to generate personalized realistic values conditioning on an individual's demographic characteristics and treatments. Our proposed model, IGNITE (Individualized GeNeration of Imputations in Time-series Electronic health records), utilises a conditional dual-variational autoencoder augmented with dual-stage attention to generate missing values for an individual. In IGNITE, we further propose a novel individualized missingness mask (IMM), which helps our model generate values based on the individual's observed data and missingness patterns. We further extend the use of IGNITE from imputing missingness to a personalized data synthesizer, where it generates missing EHRs that were never observed prior or even generates new patients for various applications. We validate our model on three large publicly available datasets and show that IGNITE outperforms state-of-the-art approaches in missing data reconstruction and task prediction.

Few-shot learning is a promising approach to molecular property prediction as supervised data is often very limited. However, many important molecular properties depend on complex molecular characteristics -- such as the various 3D geometries a molecule may adopt or the types of chemical interactions it can form -- that are not explicitly encoded in the feature space and must be approximated from low amounts of data. Learning these characteristics can be difficult, especially for few-shot learning algorithms that are designed for fast adaptation to new tasks. In this work, we develop molecular embeddings that encode complex molecular characteristics to improve the performance of few-shot molecular property prediction. Our approach leverages large amounts of synthetic data, namely the results of molecular docking calculations, and a multi-task learning paradigm to structure the embedding space. On multiple molecular property prediction benchmarks, training from the embedding space substantially improves Multi-Task, MAML, and Prototypical Network few-shot learning performance. Our code is available at https://github.com/cfifty/IGNITE.

Check out all the on-demand sessions from the Intelligent Security Summit here. The metaverse is becoming one of the hottest topics not only in technology but in the social and economic spheres. Tech giants and startups alike are already working on creating services for this new digital reality. The metaverse is slowly evolving into a mainstream virtual world where you can work, learn, shop, be entertained and interact with others in ways never before possible. Gartner recently listed the metaverse as one of the top strategic technology trends for 2023, and predicts that by 2026, 25% of the population will spend at least one hour a day there for work, shopping, education, social activities and/or entertainment.

PyTorch Adapt is a library for domain adaptation, a type of machine learning algorithm that re-purposes existing models to work in new domains. It is a fully-featured toolkit, allowing users to create a complete train/test pipeline in a few lines of code. It is also modular, so users can import just the parts they need, and not worry about being locked into a framework. One defining feature of this library is its customizability. In particular, complex training algorithms can be easily modified and combined, thanks to a system of composable, lazily-evaluated hooks. In this technical report, we explain in detail these features and the overall design of the library.

At Microsoft, we believe that energized, empowered employees are the key to a durable, competitive advantage for every organization. The Microsoft Work Trend Index shows that leaders today need to end productivity paranoia, embrace the fact that people come into the office for each other, and re-recruit everyone.1 Empowering today's digitally connected, distributed workforce requires the right culture and the right technology. At Microsoft Ignite, we're sharing new innovations across Microsoft 365, Microsoft Teams, and Microsoft Viva to help everyone thrive. Global experiences, localized content, in-person opportunities, and more--let's get ready for a new kind of Microsoft Ignite. Microsoft 365 is the cloud-first platform for all the ways that people work today--wherever, whenever, however.

Microsoft unveils big-data-capable SQL Server 2019 and extended AI capabilities to power data-driven innovation. Microsoft CEO Satya Nadella set the tone at the September 24-27 Ignite events in Orlando by sharing at least half a dozen stories of leading companies innovating and pioneering new business models with the aid of artificial intelligence (AI). It was a crisp, one-hour presentation long on vision and surprisingly short on promotion or even mentions of the significant technology announcements that followed. Nadella warned the more than 30,000 attendees that the ability to innovate and drive new business models is as much or more about changing corporate cultures and business processes as it is about applying technology. And when the technology decisions are ready to be made, Nadella counselled executives to know which capabilities are commodities and which warrant custom development to drive differentiation.

With ignite, you can write loops to train the network in just a few lines, add standard metrics calculation out of the box, save the model, etc. Well, for those who have moved from TF to PyTorch, we can say that the ignite – Keras library for PyTorch. I will not spend time talking about how cool the framework PyTorch is. Anyone who has already used it understands what I am writing about. But, with all its advantages, it is still low-level in terms of writing loops for training, checking, testing neural networks.

But here is the problem. Only a third of developers seem to know how to test these systems. And many companies have capabilities to only test them partially, risking the reliability of the system as a whole. We've decided to develop a way to know when AI algorithms work and when they don't. While it may not be critical if a movie recommendation is not that accurate, the results can be devastating if an algorithm performs poorly in an autonomous car or a medical app.