We present a novel approach called Optimized Directed Roadmap Graph (ODRM). It is a method to build a directed roadmap graph that allows for collision avoidance in multi-robot navigation. This is a highly relevant problem, for example for industrial autonomous guided vehicles. The core idea of ODRM is, that a directed roadmap can encode inherent properties of the environment which are useful when agents have to avoid each other in that same environment. Like Probabilistic Roadmaps (PRMs), ODRM's first step is generating samples from C-space. In a second step, ODRM optimizes vertex positions and edge directions by Stochastic Gradient Descent (SGD). This leads to emergent properties like edges parallel to walls and patterns similar to two-lane streets or roundabouts. Agents can then navigate on this graph by searching their path independently and solving occurring agent-agent collisions at run-time. Using the graphs generated by ODRM compared to a non-optimized graph significantly fewer agent-agent collisions happen. We evaluate our roadmap with both, centralized and decentralized planners. Our experiments show that with ODRM even a simple centralized planner can solve problems with high numbers of agents that other multi-agent planners can not solve. Additionally, we use simulated robots with decentralized planners and online collision avoidance to show how agents are a lot faster on our roadmap than on standard grid maps.

We study a security threat to reinforcement learning where an attacker poisons the learning environment to force the agent into executing a target policy chosen by the attacker. As a victim, we consider RL agents whose objective is to find a policy that maximizes average reward in undiscounted infinite-horizon problem settings. The attacker can manipulate the rewards or the transition dynamics in the learning environment at training-time and is interested in doing so in a stealthy manner. We propose an optimization framework for finding an \emph{optimal stealthy attack} for different measures of attack cost. We provide sufficient technical conditions under which the attack is feasible and provide lower/upper bounds on the attack cost. We instantiate our attacks in two settings: (i) an \emph{offline} setting where the agent is doing planning in the poisoned environment, and (ii) an \emph{online} setting where the agent is learning a policy using a regret-minimization framework with poisoned feedback. Our results show that the attacker can easily succeed in teaching any target policy to the victim under mild conditions and highlight a significant security threat to reinforcement learning agents in practice.

Legal technology is currently receiving a lot of attention from various angles. In this contribution we describe the main technical components of a system that is currently under development in the European innovation project Lynx, which includes partners from industry and research. The key contribution of this paper is a workflow manager that enables the flexible orchestration of workflows based on a portfolio of Natural Language Processing and Content Curation services as well as a Multilingual Legal Knowledge Graph that contains semantic information and meaningful references to legal documents. We also describe different use cases with which we experiment and develop prototypical solutions.

Medical Triage is of paramount importance to healthcare systems, allowing for the correct orientation of patients and allocation of the necessary resources to treat them adequately. While reliable decision-tree methods exist to triage patients based on their presentation, those trees implicitly require human inference and are not immediately applicable in a fully automated setting. On the other hand, learning triage policies directly from experts may correct for some of the limitations of hard-coded decision-trees. In this work, we present a Deep Reinforcement Learning approach (a variant of DeepQ-Learning) to triage patients using curated clinical vignettes. The dataset, consisting of 1374 clinical vignettes, was created by medical doctors to represent real-life cases. Each vignette is associated with an average of 3.8 expert triage decisions given by medical doctors relying solely on medical history. We show that this approach is on a par with human performance, yielding safe triage decisions in 94% of cases, and matching expert decisions in 85% of cases. The trained agent learns when to stop asking questions, acquires optimized decision policies requiring less evidence than supervised approaches, and adapts to the novelty of a situation by asking for more information. Overall, we demonstrate that a Deep Reinforcement Learning approach can learn effective medical triage policies directly from expert decisions, without requiring expert knowledge engineering. This approach is scalable and can be deployed in healthcare settings or geographical regions with distinct triage specifications, or where trained experts are scarce, to improve decision making in the early stage of care.

In this paper, we proposed two modified neural network architectures based on SFANet and SegNet respectively for accurate and efficient crowd counting. Inspired by SFANet, the first model is attached with two novel multi-scale-aware modules, called ASSP and CAN. This model is called M-SFANet. The encoder of M-SFANet is enhanced with ASSP containing parallel atrous convolution with different sampling rates and hence able to extract multi-scale features of the target object and incorporate larger context. To further deal with scale variation throughout an input image, we leverage contextual module called CAN which adaptively encodes the scales of the contextual information. The combination yields an effective model for counting in both dense and sparse crowd scenes. Based on the SFANet's decoder structure, M-SFANet's decoder has dual paths, for density map generation and attention map generation. The second model is called M-SegNet. For M-SegNet, we simply change bilinear upsampling used in SFANet to max unpooling originally from SegNet and propose the faster model while providing competitive counting performance. Designed for high-speed surveillance applications, M-SegNet has no additional multi-scale-aware module in order to not increase the complexity. Both models are encoder-decoder based architectures and end-to-end trainable. We also conduct extensive experiments on four crowd counting datasets and one vehicle counting dataset to show that these modifications yield algorithms that could outperform some state-of-the-art crowd counting methods.

Any motor vehicle enthusiast will tell you, with uncanny accuracy, the specs of the latest luxury rides. They will verbally pick a Lamborghini apart and elucidate on its superior qualities. Collaborative robot market is projected to grow at CAGR of 41.8%

A preprint paper coauthored by Uber AI scientists and Jeff Clune, a research team leader at San Francisco startup OpenAI, describes Fiber, an AI development and distributed training platform for methods including reinforcement learning (which spurs AI agents to complete goals via rewards) and population-based learning. The team says that Fiber expands the accessibility of large-scale parallel computation without the need for specialized hardware or equipment, enabling non-experts to reap the benefits of genetic algorithms in which populations of agents evolve rather than individual members. Fiber -- which was developed to power large-scale parallel scientific computation projects like POET -- is available in open source as of this week, on Github. It supports Linux systems running Python 3.6 and up and Kubernetes running on public cloud environments like Google Cloud, and the research team says that it can scale to hundreds or even thousands of machines. As the researchers point out, increasing computation underlies many recent advances in machine learning, with more and more algorithms relying on distributed training for processing an enormous amount of data.

Which is better, a lead foot on the brakes and a light-foot on the gas, or a lead foot on the gas and a featherweight foot on the brakes? If you are trying to drive onto the freeway, you usually need to double down on the gas pedal and make sure you enter into traffic at a fast and equitable speed. If you are driving in a busy mall parking lot, probably best to keep your foot leaning on the brakes so that you don't hit anyone. I remember when I was guiding my children on how to drive a car that it seemed like they would inevitably drift toward having a heavy foot and a light foot on each of the respective pedals. Over time, they became proficient in judging how much pressure to apply for the gas and the brakes, doing so as based on the situation and the nature of the driving circumstances involved. Today, they put little conscious thought into the matter and are seasoned drivers. Novice drivers though aren't quite sure how to treat the car controls. Besides my own children, I've seen the teenagers of other parents that were also apt to misjudge the controls when first learning to drive. It was somewhat comical one day to watch as a teenager drove down our street and his car seemed to start and stop. One moment the accelerator was being pushed, the next moment the teenager plied on the brakes. This makes sense in that he was concerned once his momentum got going that he was perhaps barreling too fast, so he wanted to slow down, but once he slowed down it became apparent that he needed to add some gas to get going again. I'm sure we've all had the same experience when trying to learn to drive. I'd also bet that sometimes you've found yourself thrown a kilter when trying to drive someone else's car, and you were unsure of how sensitive the car controls were. Whenever I rent a car, which I do a lot of the time due to my work travel, I often discover that during the first few minutes of driving the rental car that I am over-controlling it. I need to initially get used to how the brakes react, how the accelerator reacts, and how the steering reacts. It does though give you a pause for thought and perhaps allow you to reminisce about the old days of when you first learned to drive.

This research aims to help scientists develop a better understanding of COVID-19. They will use the data to understand how fast the virus is spreading in specific areas, identify high risk areas in the country, and to identify who is most at risk by better understanding symptoms linked to health conditions. The app gives researchers an opportunity see how symptoms evolve over time in different risk groups and to find patterns to who gets a mild disease. This information will be very important if there is second wave of the virus late this year or next year. The app has launched in the UK and 1.3 million people are already logging their symptoms.

If you have any questions, ask me. Does this sound familiar to you? And in case you were wondering, yes, these are a chatbot's words. So what exactly is a chatbot? Chatbots are a kind of a program that uses AI, or artificial intelligence, to comprehend and accordingly respond to the written or spoken word.