Collaborating Authors

IBM takes the next step with Cloud Pak for Data


Ever since IBM unveiled Cloud Pak for Data as a cloud-native integrated set of analytics and AI platform, we've been wondering when IBM would take the next step and announce a full-blown managed cloud service. It's now starting to happen as IBM is rolling out IBM Cloud Pak for Data as a Service. Roll back the tape to last spring when we reviewed IBM Cloud Satellite; we noted that IBM's primary cloud message has been about multi-cloud, or at least cloud-agnostic. Propelled by Red Hat OpenShift, IBM carved out such a strategy for this managed Kubernetes environment where you could deploy open source software yourself on the hardware or public cloud of your choice or choose IBM to run a managed OpenShift service for you in the IBM Cloud. That is now getting repeated with Cloud Pak for Data.

[R] NeurIPS 2020 (Spotlight) Self-Supervised Relational Reasoning for Representation Learning


I would like to share paper/code of our latest work entitled "Self-Supervised Relational Reasoning for Representation Learning" that has been accepted at NeurIPS 2020. There are three key technical differences with contrastive methods like SimCLR: (i) the replacement of the reprojection head with a relation module, (ii) the use of a Binary Cross Entropy loss (BCE) instead of a contrastive loss, and (iii) the use of multiple augmentations instead of just two. In the GitHub repository we have also released some pretrained models, minimalistic code of the method, a step-by-step notebook, and code to reproduce the experiments. Abstract: In self-supervised learning, a system is tasked with achieving a surrogate objective by defining alternative targets on a set of unlabeled data. The aim is to build useful representations that can be used in downstream tasks, without costly manual annotation.

Interpretability in Machine Learning: An Overview


This essay provides a broad overview of the sub-field of machine learning interpretability. While not exhaustive, my goal is to review conceptual frameworks, existing research, and future directions. I follow the categorizations used in Lipton et al.'s Mythos of Model Interpretability, which I think is the best paper for understanding the different definitions of interpretability. We'll go over many ways to formalize what "interpretability" means. Broadly, interpretability focuses on the how. It's focused on getting some notion of an explanation for the decisions made by our models. Below, each section is operationalized by a concrete question we can ask of our machine learning model using a specific definition of interpretability. If you're new to all this, we'll first briefly explain why we might care about interpretability at all.

Leon Eisen, PhD on LinkedIn: Mark Minevich - Wonderland AI Summit 2020


'AI can make the world better, safer, healthier, with more social innovations, not just for a few of us, but for ALL OF US!' Amazing Mark Minevich, President of Going Global Ventures and one of the biggest AI enthusiasts we ever met, gave a great talk to our virtual audience at Wonderland AI Summit 2020. If you missed Mark's presentation, here is a chance to find out how the future and society look like from Mark's perspective.

Automated driving system assessment - Siemens partners with IVEX


It will take some time before we can carelessly read the newspaper in the back seat of our self-driving car. Nevertheless, the automotive industry is working hard to push the limits in the development of vehicles with higher levels of autonomy. One of the major challenges the industry is facing is how to test an automated driving system. They also need to validate that autonomous vehicles are safe enough to be released on the public road. The verification and validation (V&V) process of automated driving systems is a challenging task, requiring a complex setup of tests.

Govzilla Announces New Customer Platform and Company Name Change to Redica Systems


Govzilla, the regulatory and compliance platform that provides data and insights to quality and safety professionals in regulated industries globally, has announced that it is rolling out an entirely redesigned customer platform. The new Enforcement Analytics platform provides customers with more data, more powerful analytics and insights, and a significantly improved user experience. It builds on the company's legacy as the largest regulatory enforcement and inspection database in the world, vastly improving its artificial intelligence models and marking a significant shift in how companies approach compliance decision-making. The launch of the new platform coincides with the rollout of a new corporate brand and identity. The company will begin operating under the name Redica Systems, effective immediately.

Artificial Intelligence is now Capable of Designing Medical Websites


Artificial intelligence (AI) coding can be used to improve medical websites in various ways, from custom personalised content presentations to the integration of unique medical AI features. These include medical appointment scheduling software, healthcare cost estimators, design medical website, prescribe medication, and answer questions. By developing more targeted and custom AI solutions, some artificial intelligence healthcare platforms aim to usher in the more widespread adoption of medical AI technologies, benefitting organisations, practices, clinicians, and patients alike. AI medical websites can use AI tools to present targeted information specific to the consumer. AI medical websites use IP addresses to locate the user and present information specific to physicians and practices local to their area.

Key skills to Transition into Artificial Intelligence in 2020


Artificial Intelligence and Machine Learning are the two trending technologies managing the current market place. These two can change how organizations work and people interact with one another to perform complex tasks. However, the issues that AI solves are difficult and to work in the AI industry you will require a solid and focused set of skills. Before we go to realize the precise skills needed to progress into AI. Let's see how businesses are receiving this innovation to perform the different assignments in a better and simple way. Let's have a look at the Adoption of this technology in the industries Things considered, as the tide of AI and ML keeps on creating.

This Squishy 3D-Printed Human Heart Feels Like the Real Thing


In the intro to the HBO sci-fi series Westworld, a 3D printer churns out humanoid robots, delicately assembling the incredible complexities of the human form so that those robots can go on to--spoiler alert--do naughty things. It takes a lot of biomechanical coordination, after all, to murder a whole lot of flesh-and-blood people. Speaking of: Researchers just made a scientific leap toward making 3D-printed flesh and blood a reality. Writing recently in the journal ACS Biomaterials Science & Engineering, a team described how they repurposed a low-cost 3D printer into one capable of turning an MRI scan of a human heart into a deformable full-size analog you can actually hold in your hand. Squeeze it, and it'll give like the real thing.

Can collision-aversion make motorcycles safer?


My motorcycle accident was a classic, the scenario you hear about in rider safety programs and read about on forums. I was cruising down a four-lane city street with no traffic in my direction but bumper-to-bumper gridlock in the oncoming lanes. At a dogleg in the road I rounded a corner to find a car from one of those oncoming lanes turning left over the double yellow into a gas station parking lot. It was textbook, something I realized even as it was happening. My motorcycle slammed into the front fender of the turning car and I came off the bike, landing on the sidewalk 15 feet away. If not for an airbag vest I wore religiously--an inflatable powered by a CO2 cartridge and clipped to the bike's frame via a tether, which acts like a rip cord when rider and machine are parted--I'm convinced I might not be writing this.