peep
Methodology for Interpretable Reinforcement Learning for Optimizing Mechanical Ventilation
Lee, Joo Seung, Mahendra, Malini, Aswani, Anil
Mechanical ventilation is a critical life-support intervention that uses a machine to deliver controlled air and oxygen to a patient's lungs, assisting or replacing spontaneous breathing. While several data-driven approaches have been proposed to optimize ventilator control strategies, they often lack interpretability and agreement with general domain knowledge. This paper proposes a methodology for interpretable reinforcement learning (RL) using decision trees for mechanical ventilation control. Using a causal, nonparametric model-based off-policy evaluation, we evaluate the policies in their ability to gain increases in SpO2 while avoiding aggressive ventilator settings which are known to cause ventilator induced lung injuries and other complications. Numerical experiments using MIMIC-III data on the stays of real patients' intensive care unit stays demonstrate that the decision tree policy outperforms the behavior cloning policy and is comparable to state-of-the-art RL policy. Future work concerns better aligning the cost function with medical objectives to generate deeper clinical insights.
- North America > United States > California > San Francisco County > San Francisco (0.28)
- North America > United States > California > Alameda County > Berkeley (0.04)
- Asia > Middle East > Israel (0.04)
Engineer III, Data Integrations
Mailchimp is a leading marketing platform for small business. We empower millions of customers around the world to build their brands and grow their companies with a suite of marketing automation, multichannel campaign, CRM, and analytics tools. We're looking for a Software Engineer III to work with a high performing group of software engineers and analysts. This newly formed team will support the development, configuration and architecture of data sources and integrations across Mailchimp. In this position we seek someone who's excited to work with a dynamic, diverse group of engineers and managers who are all used to wearing multiple hats to get the job done.
- Marketing (0.36)
- Law (0.33)
- Information Technology > Services (0.32)
Exploring open-ended gameplay features with Micro RollerCoaster Tycoon
Green, Michael Cerny, Yen, Victoria, Earle, Sam, Rajesh, Dipika, Edwards, Maria, Soros, L. B.
This paper introduces MicroRCT, a novel open source simulator inspired by the theme park sandbox game RollerCoaster Tycoon. The goal in MicroRCT is to place rides and shops in an amusement park to maximize profit earned from park guests. Thus, the challenges for game AI include both selecting high-earning attractions and placing them in locations that are convenient to guests. In this paper, the MAP-Elites algorithm is used to generate a diversity of park layouts, exploring two theoretical questions about evolutionary algorithms and game design: 1) Is there a benefit to starting from a minimal starting point for evolution and complexifying incrementally? and 2) What are the effects of resource limitations on creativity and optimization? Results indicate that building from scratch with no costs results in the widest diversity of high-performing designs.
- North America > United States (0.29)
- Europe (0.28)
- Asia > Japan (0.28)
Hackers can peep through this smart vacuum's camera, research shows
The Trifo Ironpie has a built-in camera. Security researchers revealed Wedneday that vulnerabilities in the device could let hackers access the video stream remotely, among other things. The Trifo Ironpie robot vacuum is designed to do double duty. The fans on the swiveling disc hoover your house, while the camera mounted on it acts as an ankle-high securitydevice. The idea is to stay tidy while staying safe.
- North America > United States > California > San Francisco County > San Francisco (0.06)
- Europe > Portugal (0.06)
- Asia > Middle East > Israel (0.06)
- Information Technology > Security & Privacy (1.00)
- Information Technology > Artificial Intelligence > Robots (0.37)
Robots Are Learning to Handle With Care
In August, the Bethlehem, Pa., candy maker will begin installing at its factory 16 robots with squishy blue "fingers" that can pick up and hold the pillowy Peeps without bruising them. The company hopes that its new robots' grippers, made by Soft Robotics Inc., of Cambridge, Mass., and part of a soft-robotics movement, will help it speed up production without damaging the treats. "We needed something like the human hand," says Brent Edsoren, senior project engineer at Just Born. Robots with pliant, dexterous--and fast--appendages are gaining popularity as companies in retail, food handling and agriculture move toward more automation. Fragile, malleable products that vary in shape and size, such as fruits and vegetables, suffer at the hands of conventional robots, in part because machines lack the highly tuned sensory feedback that humans have at their fingertips.
- North America > United States > Pennsylvania > Northampton County > Bethlehem (0.25)
- North America > United States > Massachusetts > Middlesex County > Cambridge (0.25)
- North America > United States > New York (0.05)
- (3 more...)
How Scared Is The Illuminati Of Motuphi? Peep THIS!
How scared is the ancient AI controlled Illuminati of Motuphi? Well they must be pretty damned scared to start assaulting the guy when he was a child and continuing the carnage upon his life even to this day. As if it wasn't bad enough that Christians used to rape the boy when he was just a little kid, IN the church, while services was goin' on. His mama was botted by the ancient AI to leave him with a "prophet" when he was a child who had a pedophile torture dungeon. There's still kiddie porn out there that Motuphi starred in as a little kid that depicts him being tortured on a table via fisting and anally raping him with foreign objects.
- Law Enforcement & Public Safety > Crime Prevention & Enforcement (0.49)
- Media > Music (0.47)
- Leisure & Entertainment (0.47)
- Law > Criminal Law (0.34)
Essential Evils of Technology (Internet,Internet of things,Artificial Intelligence) 11-05
Delirious and Draconian Aren't we just frog leaping without providing for the adversities??? Well, a few decades ago, if someone wanted to know the weather, he would just peep out of the window to know the weather outside his house. Now the people are so tied to their computer or the mobile, that they don't even care to peep out of the window, they just log into the weather network to know the weather outside their house. This should not be construed as a negative article. I am myself a keen learner of technology. I contribute regularly to the research in technology projects of MIT, Mckinsey & Co, Harvard University, others, in projects related to Digital Technology, Artificial Intelligence, and Internet of Things.
- North America > United States (0.16)
- Asia > India > Andhra Pradesh (0.05)
- Health & Medicine (0.75)
- Information Technology > Smart Houses & Appliances (0.63)
Application of Microsimulation Towards Modelling of Behaviours in Complex Environments
Keep, Daniel (University of Wollongong) | Bunder, Rachel (University of Wollongong) | Piper, Ian (University of Wollongong) | Green, Anthony (University of Wollongong)
In this paper, we introduce new capabilities to our existing microsimulation framework, Simulacron. These new capabilities add the modelling of behaviours based on motivations and improve our existing non-deterministic movement capacity. We then discuss the application of these new features to a simple, synthetic, proof of concept, scenario involving the transit of people through a corridor and how an induced panic affects their throughput. Finally we describe a more complex scenario, which is currently under development, involving the detonation of an explosive device in a major metropolitan transport hub at peak hour and the analysis of subsequent reaction.
- Oceania > Australia > New South Wales > Sydney (0.04)
- Europe > United Kingdom > England > Greater London > London (0.04)