Federated learning (FL) suffers from data heterogeneity, where the diverse data distributions across clients make it challenging to train a single global model effectively. Existing personalization approaches aim to address the data heterogeneity issue by creating a personalized model for each client from the global model that fits their local data distribution. However, these personalized models may achieve lower accuracy than the global model in some clients, resulting in limited performance improvement compared to that without personalization. To overcome this limitation, we propose a per-instance personalization FL algorithm Flow. Flow creates dynamic personalized models that are adaptive not only to each client's data distributions but also to each client's data instances. The personalized model allows each instance to dynamically determine whether it prefers the local parameters or its global counterpart to make correct predictions, thereby improving clients'accuracy. We provide theoretical analysis on the convergence of Flow and empirically demonstrate the superiority of Flow in improving clients' accuracy compared to state-of-the-art personalization approaches on both vision and language-based tasks.

Normalizing flows are a class of probabilistic generative models which allow for both fast density computation and efficient sampling and are effective at modelling complex distributions like images. A drawback among current methods is their significant training cost, sometimes requiring months of GPU training time to achieve state-of-the-art results. This paper introduces Wavelet Flow, a multi-scale, normalizing flow architecture based on wavelets. A Wavelet Flow has an explicit representation of signal scale that inherently includes models of lower resolution signals and conditional generation of higher resolution signals, i.e., super resolution. A major advantage of Wavelet Flow is the ability to construct generative models for high resolution data (e.g., 1024 1024 images) that are impractical with previous models. Furthermore, Wavelet Flow is competitive with previous normalizing flows in terms of bits per dimension on standard (low resolution) benchmarks while being up to 15 faster to train.

We introduce manifold-learning flows (ℳ-flows), a new class of generative models that simultaneously learn the data manifold as well as a tractable probability density on that manifold. Combining aspects of normalizing flows, GANs, autoencoders, and energy-based models, they have the potential to represent data sets with a manifold structure more faithfully and provide handles on dimensionality reduction, denoising, and out-of-distribution detection. We argue why such models should not be trained by maximum likelihood alone and present a new training algorithm that separates manifold and density updates. In a range of experiments we demonstrate how ℳ-flows learn the data manifold and allow for better inference than standard flows in the ambient data space.

Originally published on Towards AI the World's Leading AI and Technology News and Media Company. If you are building an AI-related product or service, we invite you to consider becoming an AI sponsor. At Towards AI, we help scale AI and technology startups. Let us help you unleash your technology to the masses. In this article, we will look at how you can combine the text generation capabilities of GPT-3 with the creative image generation part of DALL.E to produce a piece of art that would have required days if not months, with the conventional setup Without further ado, let's write a poem on unstructured data in the style of Shakespear using GPT3TextGeneration Executor and generate the illustrations for the same using DALL.E-Flow.

If you spend more on coffee than IT security, you will be hacked! Warned U.S. Cybersecurity Czar Richard Clarke, speaking at RSA Conference. This quote would make a great bumper sticker if it weren't for network attacks. According to research by IBM, it takes 280 days to find and contain the average cyberattack, while the average attack costs $3.86 million. But what are network attacks, and how can we leverage a next-gen search tool like Jina to mitigate our exposure to the threat?

With neural search seeing rapid adoption, more people are looking at using it for indexing and searching through their unstructured data. I know several folks already building PDF search engines powered by AI, so I figured I'd give it a stab too. How hard could it possibly be? This is just a rough and ready roadmap -- so stay tuned to see how things really pan out. If you want to follow along at home (and maybe fix a few of my bugs!), check the repo: I want to build a search engine for a dataset of arbitrary PDFs.

Flow is a psychological state that is reported to improve people's performance. Flow can emerge when the person's skills and the challenges of their activity match. This paper applies this concept to artificial intelligence agents. We equip a decision-making agent with a metacontrol policy that guides the agent to activities where the agent's skills match the activity difficulty. Consequently, we expect the agent's performance to improve. We implement and evaluate this approach in the role-playing game of Angband.

More and more producers are looking into solutions to improve the subtitling process. Automatic subtitling is the key to control the cost and the delay incurred by an otherwise manual subtitling process. Because expensive subtitles (or the lack thereof) may hamper the publication of valuable content, we took the challenge to radically automate the process. The Limecraft subtitling service is currently available in private beta. Producers of audiovisual content, including Film, Television and Corporate Video, consider subtitling an essential part of the production process.