Welcome to VMware AI & Advanced Services! Below you’ll find a showcase of diverse and innovative AI/ML projects developed by our talented Research and Development team at VMware. Each project demonstrates our dedication to pushing the boundaries of AI and ML technologies and solving real-world challenges.
Below is a list of featured AI & ML open-source projects we are very excited to share with you. Join our open-source community: explore, experiment, ask questions, and contribute.
Projects | Description | Category | Originated or Contributing |
---|---|---|---|
Certifier Framework for Confidential Computing | The Certifier Framework for Confidential Computing consists of a client API called the Certifier API and server-based policy evaluation server called the Certifier Service. The Certifier API greatly simplifies and unifies programming and operations support for multi-vendor Confidential Computing platforms by providing simple client trust management, including attestation evaluation, secure storage, platform initialization, secret sharing, secure channels and other services. The Certifier Service provides support for scalable, policy-driven trust management, including attestation evaluation, application upgrade and other Confidential Computing Trust services. |
Trust/Security | Originated |
Efficient-multiclass-classification | Duet is a decision tree ensemble method based multiclass classification framework that offers a more efficient resource usage while preserving and even improving the classification accuracy in comparison to standard monolithic models. Duet is based on a small bagging ensemble model and a booting model. The current implementation of Duet is based on Random Forest and XGBoost. | Resource Efficient Classification | Originated |
Efficient-supervised-anomaly-detection | RADE is a resource-efficient decision tree ensemble method (DTEM) based anomaly detection approach that augments standard DTEM classifiers resulting in competitive anomaly detection capabilities and significant savings in resource usage. The current implementation of RADE augments either Random-Forest or XGBoost. |
Resource Efficient Classification | Originated |
RAIL | Helpful tidbits to advance the open source NLP community, published models on https://huggingface.co/VMware Access our fine tuned models on hugging face |
LLM | Originated |
srrcomp |
srrcomp (stands for structured random rotation-based compression) is a Python package that provides structured random rotation-based compression techniques with strong theoretical guarantees. In particular, srrcomp can be used for: • Fast and efficient lossy compression. • Unbiased estimates. • Distributed mean estimation. • Compressing gradient updates in distributed and federated learning. The implementation is torch-based and thus supports CPU and GPU. Part of srrcomp code has been contributed to Intel's OpenFL project and implemented by Google Research in TensorFlow Federated (TFF) project. |
Compression Techniques | Originated |
vSphere-machine-learning-extension | This project hosts codes and documents to enable machine learning workloads better running on VMware vSphere and VMware Cloud. For now, this project focuses on [Kubeflow] (https://www.kubeflow.org/). We will further extend the scope to other machine learning software soon. | Machine Learning | Originated |
Adversarial Robustness Toolbox (ART) | A Python library for machine learning security that enables ML practitioners to defend and evaluate machine learning models and applications against the adversarial threats of evasion, poisoning, extraction, and inference. | Trust/Security | Contributing |
FATE | The FATE open source project holds strategic importance for VMware, serving as a pivotal framework to enable collaborative machine learning across distributed data sources. By leveraging FATE, VMware can leverage the potential of federated learning to create advanced and ethical AI solutions that resonate with the needs of their clients. | FATE | Contributing |
FATE-Operator | The FATE-Operator enables users to seamlessly execute federated workloads using FATE on diverse Kubernetes clusters. This flexibility empowers organizations to harness the power of federated learning across their infrastructure, and collaborative AI model training. By simplifying the deployment process, FATE-Operator contributes to democratizing federated learning adoption and accelerating innovation in privacy-preserving AI solutions. | FATE | Contributing |
Kubeflow | A machine learning operations (MLOps) platform built on top of Kubernetes providing a comprehensive set of tools to streamline the deployment of machine learning workflows. | Machine Learning | Contributing |
Ray | Adds VMware platforms support, such as vSphere, to Ray Auroscaler | Machine Learning | Contributing |
At VMware AI & Advanced Services, we are passionate about harnessing the power of AI and ML to unlock new realms of possibilities. Our team of dedicated developers and thinkers combine years of experience with cutting-edge technology to create solutions that enchant and inspire.
GPL Commitment
Before filing or continuing to prosecute any legal proceeding or claim (other than a Defensive Action) arising from termination of a Covered License, VMware commits to extend to the person or entity ("you") accused of violating the Covered License the following provisions regarding cure and reinstatement, taken from GPL version 3. As used here, the term 'this License' refers to the specific Covered License being enforced.
However, if you cease all violation of this License, then your license from a particular copyright holder is reinstated (a) provisionally, unless and until the copyright holder explicitly and finally terminates your license, and (b) permanently, if the copyright holder fails to notify you of the violation by some reasonable means prior to 60 days after the cessation.
Moreover, your license from a particular copyright holder is reinstated permanently if the copyright holder notifies you of the violation by some reasonable means, this is the first time you have received notice of violation of this License (for any work) from that copyright holder, and you cure the violation prior to 30 days after your receipt of the notice.
VMware intends this Commitment to be irrevocable, and binding and enforceable.
'Covered License' means the GNU General Public License, version 2 (GPLv2), the GNU Lesser General Public License, version 2.1 (LGPLv2.1), or the GNU Library General Public License, version 2 (LGPLv2), all as published by the Free Software Foundation.
'Defensive Action' means a legal proceeding or claim that VMware brings against you in response to a prior proceeding or claim initiated by you or your affiliate.