AI Research News Update: Issue 2 (Nov 22-30, 2021)
Asif Razzaq
AI Research Editor | CEO @ Marktechpost | 1 Million Monthly Readers and 80k+ ML Subreddit Members
Researchers At Trinity College Dublin and University Of Bath Introduce A Deep Neural Network-Based Model To Improve The Quality Of Animations Containing Quadruped Animals
It’s difficult to make realistic quadruped animations. Producing realistic animations with key-framing and other techniques takes a long time and demand a lot of artistic skill. Motion capture methods, on the other hand, have their own set of obstacles (bringing the animal into the studio, attaching motion capture markers, and getting the animal to perform the intended performance), and the final animation will almost certainly require cleanup. It would be beneficial if an animator could supply a rough animation and then be given a high-quality realistic one in exchange.
Researchers at the University of Bath and Trinity College Dublin have developed a deep neural network-based technique that could help improve animations containing quadruped animals such as dogs. The team found that given an initial animation that may lack subtle details of true quadruped motion and/or contains small errors, a neural network can learn how to add these subtleties and correct errors to produce an enhanced animation while preserving the semantics and context of the original animation.
Microsoft Researchers Unlock New Avenues In Image-Generation Research With Manifold Matching Via Metric Learning
By developing fresh images, generative image models provide a distinct value. These photos could be clear super-resolution copies of current images or even manufactured shots that look realistic. The framework of training two networks against each other has shown pioneering success with Generative Adversarial Networks (GANs) and their variants: a generator network learns to generate realistic fake data that can fool a discriminator network, and the discriminator network learns to correctly tell apart the generated counterfeit data from the actual data.
The research community must address two issues in order to use the most recent advances in computer vision for GANs. First, rather than using geometric metrics, GANs model data distributions using statistical measures such as the mean and moments. Second, in classic GANs, the discriminator network loss is solely represented as a 1D scalar value corresponding to the Euclidean distance between the genuine and fake data distributions. The research community has been unable to utilize breakthrough metric learning methods directly or experiment with novel loss functions and training strategies to continue to develop generative models due to these two limitations.
Google AI Improves The Performance Of Smart Text Selection Models By Using Federated Learning
Smart Text Selection is one of Android’s most popular features, assisting users in selecting, copying, and using text by anticipating the desired word or combination of words around a user’s tap and expanding the selection appropriately. Selections are automatically extended with this feature, and users are offered an app to open selections with defined classification categories, such as addresses and phone numbers, saving them even more time.
The Google team made efforts to improve the performance of Smart Text Selection by utilizing?federated learning?to train a neural network model responsible for user interactions while maintaining personal privacy. The research team was able to enhance the model’s selection accuracy by up to?20% on some sorts of entities thanks?to this effort, which is part of Android’s new?Private Compute Core safe environment.
NVIDIA Open-Source ‘FLARE’ (Federated Learning Application Runtime Environment), Providing A Common Computing Foundation For Federated Learning
Standard machine learning methods involve storing training data on a single machine or in a data center. Federated learning is a privacy-preserving technique that is especially useful when the training data is sparse, confidential, or less diverse.?
NVIDIA open-source?NVIDIA FLARE, which stands for Federated Learning Application Runtime Environment. It is a software development kit that enables remote parties to collaborate for developing more generalizable AI models. NVIDIA FLARE is the underlying engine in the NVIDIA Clara Train’s federated learning software, which has been utilized for diverse AI applications such as medical imaging, genetic analysis, cancer, and COVID-19 research.
Ericsson And Uppsala University Team Up To Research Air Quality Prediction Using Machine Learning And Federated learning
Statistical methods have recently been applied in various sectors, spanning from health care to customer relationship management, to analyze and forecast the behavior of a given event. The goal here is to evaluate the likelihood of an event occurring rather than predict the exact outcome. However, the path is not without bumps; getting access to the data needed to deploy machine learning algorithms is difficult for the following reasons:
Google Research Open-Sources ‘SAVi’: An Object-Centric Architecture That Extends The Slot Attention Mechanism To Videos
Multiple distinct things act as compositional building blocks that can be processed independently and recombined in humans’ understanding of the world. The foundation for high-level cognitive abilities like language, causal reasoning, arithmetic, planning, and so on is a compositional model of the universe. Therefore, it’s essential for generalizing in predictable and systematic ways. Machine learning algorithms with object-centric representations have the potential to dramatically improve sampling efficiency, resilience, generalization to new problems, and interpretability.
Unsupervised multi-object representation learning is widely used in various applications. These algorithms learn to separate and represent objects from the statistical structure of the data alone, without the requirement for supervision, by using object-centric inductive biases.
领英推荐
Apple Researchers Propose A Method For Reconstructing Training Data From Diverse Machine Learning Models By Ensemble Inversion
Model inversion (MI), where an adversary abuses access to a trained Machine Learning (ML) model in order to infer sensitive information about the model’s original training data, has gotten a lot of attention in recent years. The trained model under assault is frequently frozen during MI and used to direct the training of a generator, such as a Generative Adversarial Network, to rebuild the distribution of the model’s original training data.?
As a result, scrutiny of the capabilities of MI techniques is essential for the creation of appropriate protection techniques. Reconstruction of training data with high quality using a single model is complex. However, existing MI literature does not consider targeting many models simultaneously, which could offer the adversary extra information and viewpoints. If successful, this could result in the disclosure of original training samples, putting the privacy of dataset subjects in jeopardy if the training data contains Personally Identifiable Information.
Apple researchers have presented an ensemble inversion technique that uses a generator restricted by a set of trained models with shared subjects or entities to estimate the distribution of original training data. When compared to MI of a single ML model, this technique results in considerable improvements in the quality of the generated samples with distinguishing properties of the dataset entities. Without any dataset, high-quality results were obtained, demonstrating how using an auxiliary dataset similar to the expected training data improves the outcomes. The impact of model diversity in the ensemble is examined in-depth, and extra constraints are used to encourage sharp predictions and high activations for the rebuilt samples, resulting in more accurate training picture reconstruction.
Scientists Use A New Deep Learning Method To Add 301 Planets to Kepler’s Total Count
Deep neural networks are machine learning systems that automatically learn a task if provided with necessary data.?An artificial neural network (ANN) having numerous layers between the input and output layers is known as a deep neural network (DNN). Neural networks are made available in various shapes and sizes. However, they all include the same essential components: neurons, synapses, weights, biases, and functions.
Recently, scientists have added a total of?301 validated exoplanets?to the already existing exoplanet tally. The cluster of planets is the most recent addition to the 4,569 confirmed planets orbiting various faraway stars. This news has gotten everyone into pondering how possibly scientists could discover such a huge, all at a time. The answer lies in a Deep Neural Network called ExoMiner. The Kepler Science Office upgraded all 301 machine-validated planets to planet candidate status after discovering them via the Kepler Science Operations Center pipeline. None of this was possible before the implementation of ExoMiner.
MetaICL: A New Few-Shot Learning Method Where A Language Model Is Meta-Trained To Learn To In-Context Learn
Large language models (LMs) are capable of in-context learning, which involves conditioning on a few training examples and predicting which tokens will best complete a test input. This type of learning shows promising results because the model learns a new task solely by inference, with no parameter modifications. However, the model’s performance significantly lags behind supervised fine-tuning. In addition, the results show high variance, which can make it difficult to engineer the templates required to convert existing tasks to this format.
Researchers from Facebook AI, the University of Washington, and the Allen Institute for AI have developed Meta-training for In-Context Learning (MetaICL), a new few-shot learning meta-training paradigm. In this approach, LM is meta-trained to learn in context, conditioning on training instances to recover the task and generate predictions.
TensorFlow Introduces TensorFlow Graph Neural Networks (TF-GNNs)
In the actual world and also in engineered systems, graphs are everywhere. A graph is a representation of a collection of entities such as objects, places, or people, as well as the relationships between them. The data which is seen in machine learning problems is usually structured or relational and hence can be represented as a graph.
A graph depicts the connections (edges) between various items (nodes or vertices). Each node, edge, or full graph can be characterized, and information can be stored in each of these components. Edges can also have directionality assigned to them to define information or traffic flow, for example.
A Graph Neural Network (GNN) is a type of neural network that is used to handle data in graph data structures. GNNs can be used to answer inquiries concerning a variety of graph features. GNN tries to anticipate the properties of the complete graph by working at the graph level. Using GNN, the existence of particular “shapes” in a network, such as circles, which could indicate sub-molecules or intimate social interactions, can be detected. Similar to image classification or segmentation, GNNs can be employed on node-level tasks to categorize nodes in a network and predict partitions and affinities in a graph. Finally, GNNs can be utilized to detect connections between entities at the edge level, for example, by “pruning” edges to determine the state of objects in a scene.
Quick Read: https://www.marktechpost.com/2021/11/22/tensorflow-introduces-tensorflow-graph-neural-networks-tf-gnns/
About:?
Marktechpost?is a California-based AI News Platform providing easy-to-consume, byte size updates in machine learning, deep learning, and data science research. Our vision is to showcase the hottest research trends in AI from around the world using our innovative method of search and discovery.
Asif Razzaq:?Asif Razzaq is an AI Journalist and Cofounder of Marktechpost, LLC. He is a visionary, entrepreneur, and engineer who aspires to use the power of Artificial Intelligence for good.
For Sponsorship or Partnership Contact us at [email protected]
You can also share your AI-related research directly via LinkedIn message or email at [email protected]
Feel free to join our?37k+ AI Group on Facebook