The causes underlying unfair decision making are complex, being internalised in different ways by decision makers, other actors dealing with data and models, and ultimately by the individuals being affected by these decisions. One frequent manifestation of all these latent causes arises in the form of missing values: protected groups are more reluctant to give information that could be used against them, delicate information for some groups can be erased by human operators, or data acquisition may simply be less complete and systematic for minority groups. As a result, missing values and bias in data are two phenomena that are tightly coupled. However, most recent techniques, libraries and experimental results dealing with fairness in machine learning have simply ignored missing data. In this paper, we claim that fairness research should not miss the opportunity to deal properly with missing data. To support this claim, (1) we analyse the sources of missing data and bias, and we map the common causes, (2) we find that rows containing missing values are usually fairer than the rest, which should not be treated as the uncomfortable ugly data that different techniques and libraries get rid of at the first occasion, and (3) we study the trade-off between performance and fairness when the rows with missing values are used (either because the technique deals with them directly or by imputation methods). We end the paper with a series of recommended procedures about what to do with missing data when aiming for fair decision making.

A continual learning agent should be able to build on top of existing knowledge to learn on new data quickly while minimizing forgetting. Current intelligent systems based on neural network function approximators arguably do the opposite-- they are highly prone to forgetting and rarely trained to facilitate future learning. One reason for this poor behavior is that they learn from a representation that is not explicitly trained for these two goals. In this paper, we propose MRCL, an objective to explicitly learn representations that accelerate future learning and are robust to forgetting under online updates in continual learning. The idea is to optimize the representation such that online updates minimize error on all samples with little forgetting. We show that it is possible to learn representations that are more effective for online updating and that sparsity naturally emerges in these representations. Moreover, our method is complementary to existing continual learning strategies, like MER, which can learn more effectively from representations learned by our objective. Finally, we demonstrate that a basic online updating strategy with our learned representation is competitive with rehearsal based methods for continual learning.

Mining temporal assertions from time-series data using information theory to filter real properties from incidental ones is a practically significant challenge. The problem is complex for continuous or hybrid systems because the degrees of influence on a consequent from a timed-sequence of predicates (called its prefix sequence), varies continuously over dense time intervals. We propose a parameterized method that uses interval arithmetic for flexibly learning prefix sequences having influence on a defined consequent over various time scales and predicates over system variables.

We focus on the important problem of emergency evacuation, which clearly could benefit from reinforcement learning that has been largely unaddressed. Emergency evacuation is a complex task which is difficult to solve with reinforcement learning, since an emergency situation is highly dynamic, with a lot of changing variables and complex constraints that makes it difficult to train on. In this paper, we propose the first fire evacuation environment to train reinforcement learning agents for evacuation planning. The environment is modelled as a graph capturing the building structure. It consists of realistic features like fire spread, uncertainty and bottlenecks. We have implemented the environment in the OpenAI gym format, to facilitate future research. We also propose a new reinforcement learning approach that entails pretraining the network weights of a DQN based agents to incorporate information on the shortest path to the exit. We achieved this by using tabular Q-learning to learn the shortest path on the building model's graph. This information is transferred to the network by deliberately overfitting it on the Q-matrix. Then, the pretrained DQN model is trained on the fire evacuation environment to generate the optimal evacuation path under time varying conditions. We perform comparisons of the proposed approach with state-of-the-art reinforcement learning algorithms like PPO, VPG, SARSA, A2C and ACKTR. The results show that our method is able to outperform state-of-the-art models by a huge margin including the original DQN based models. Finally, we test our model on a large and complex real building consisting of 91 rooms, with the possibility to move to any other room, hence giving 8281 actions. We use an attention based mechanism to deal with large action spaces. Our model achieves near optimal performance on the real world emergency environment.

The incorporation of macro-actions (temporally extended actions) into multi-agent decision problems has the potential to address the curse of dimensionality associated with such decision problems. Since macro-actions last for stochastic durations, multiple agents executing decentralized policies in cooperative environments must act asynchronously. We present an algorithm that modifies generalized advantage estimation for temporally extended actions, allowing a state-of-the-art policy optimization algorithm to optimize policies in Dec-POMDPs in which agents act asynchronously. We show that our algorithm is capable of learning optimal policies in two cooperative domains, one involving real-time bus holding control and one involving wildfire fighting with unmanned aircraft. Our algorithm works by framing problems as "event-driven decision processes," which are scenarios in which the sequence and timing of actions and events are random and governed by an underlying stochastic process. In addition to optimizing policies with continuous state and action spaces, our algorithm also facilitates the use of event-driven simulators, which do not require time to be discretized into time-steps. We demonstrate the benefit of using event-driven simulation in the context of multiple agents taking asynchronous actions. We show that fixed time-step simulation risks obfuscating the sequence in which closely separated events occur, adversely affecting the policies learned. In addition, we show that arbitrarily shrinking the time-step scales poorly with the number of agents.

A link has been posted to your Facebook feed. A budding deal to combine Jeep maker Fiat Chrysler Automobiles and French automaker Renault could make that possibility a reality in the coming years. After years of searching for a partner, Fiat Chrysler may have finally found the one. Less than a year after the death of consolidation promoter and Fiat Chrysler savior Sergio Marchionne, a merger of Fiat Chrysler and Renault is on the verge of happening. Fiat Chrysler on Monday unveiled its proposal to combine the two companies in a 50-50 deal that would carry seismic consequences for the global automotive business. Here's how it could affect American car shoppers: Fiat Chrysler is badly lacking electric vehicles in America.

A'selfie robot' that is set to replace instant cameras and photo booths at parties has made its debut at at a UK wedding. The robot, designed by Birmingham-based firm, allowed wedding goers at a local reception to send photos it to themselves by email or text. Armed with AI software that can detect faces, the robot can'roam freely' around a room and stops to ask people if they would like their photo taken. It also has an infrared sensor that prevents it smashing into people and obstacles. Eva Photography Robot has been developed by a computer programmer in Birmingham to take selfies of partygoers and made its first debut at a wedding.

A startup in China will be the first company in the world to begin mass-producing self-driving delivery vehicles for some of the country's biggest commerce giants. According to a report from Bloomberg, the company Neolix has begun production on 1,000 level four autonomous vehicles that it plans to roll out in China throughout the next year. The tiny vans, which are essentially four-wheeled robots outfitted with trunks for storage, are capable of navigating their environment without any human pilot and have already garnered interest from two of China's biggest retailers: Huawei and JD.com. Neolix's robotic courier will cost around $30,000 each and could usher in a new era for e-commerce in China where companies like Alibaba have exploded in scope. In 2019 alone, Alibaba has generated about $152 billion.

The Diet passed into law Tuesday a bill allowing drivers to use their smartphones while their cars are traveling autonomously under certain circumstances and if they are able to shift to manual driving immediately during an emergency. The bill to amend the road traffic law, which includes rules on the Level 3 self-driving of vehicles under certain circumstances, passed the House of Representatives. The amended law will come into effect by May next year. In Level 3 situations, automated driving is permitted under conditions that have to do with the type of road, the vehicle speed and other factors. A Level 3 situation includes a traffic jam on an expressway.

A new type of artificial intelligence can generate a "living portrait" from just one image. The enigmatic, painted smile of the "Mona Lisa" is known around the world, but that famous face recently displayed a startling new range of expressions, courtesy of artificial intelligence (AI). In a video shared to YouTube on May 21, three video clips show disconcerting examples of the Mona Lisa as she moves her lips and turns her head. She was created by a convolutional neural network -- a type of AI that processes information much as a human brain does, to analyze and process images. Researchers trained the algorithm to understand facial features' general shapes and how they behave relative to each other, and then to apply that information to still images.