"I wasn't able to go outside anywhere alone," Ana says. "I had so much anxiety that when I went outside to do errands, I lost consciousness twice. That's when I realized I was very sick." Ana began working for LGBTQ dating app Grindr when she was in her early twenties, one of hundreds of Hondurans hired by US-headquartered outsourcing company PartnerHero to work on the account. Her team was based in San Pedro Sula, Honduras' second city, where they handled tasks ranging from the mundane--tech support emails and billing queries--to the horrifying: user reports of sexual assault, homophobic violence, child sexual abuse, and murder.

Recently, visual-language navigation (VLN) -- entailing robot agents to follow navigation instructions -- has shown great advance. However, existing literature put most emphasis on interpreting instructions into actions, only delivering "dumb" wayfinding agents. In this article, we devise LANA, a language-capable navigation agent which is able to not only execute human-written navigation commands, but also provide route descriptions to humans. This is achieved by simultaneously learning instruction following and generation with only one single model. More specifically, two encoders, respectively for route and language encoding, are built and shared by two decoders, respectively, for action prediction and instruction generation, so as to exploit cross-task knowledge and capture task-specific characteristics. Throughout pretraining and fine-tuning, both instruction following and generation are set as optimization objectives. We empirically verify that, compared with recent advanced task-specific solutions, LANA attains better performances on both instruction following and route description, with nearly half complexity. In addition, endowed with language generation capability, LANA can explain to humans its behaviors and assist human's wayfinding. This work is expected to foster future efforts towards building more trustworthy and socially-intelligent navigation robots.

How can one meaningfully make a measurement, if the meter does not conform to any standard and its scale expands or shrinks depending on what is measured? In the present work it is argued that current evaluation practices for machine-learning classifiers are affected by this kind of problem, leading to negative consequences when classifiers are put to real use; consequences that could have been avoided. It is proposed that evaluation be grounded on Decision Theory, and the implications of such foundation are explored. The main result is that every evaluation metric must be a linear combination of confusion-matrix elements, with coefficients - "utilities" - that depend on the specific classification problem. For binary classification, the space of such possible metrics is effectively two-dimensional. It is shown that popular metrics such as precision, balanced accuracy, Matthews Correlation Coefficient, Fowlkes-Mallows index, F1-measure, and Area Under the Curve are never optimal: they always give rise to an in-principle avoidable fraction of incorrect evaluations. This fraction is even larger than would be caused by the use of a decision-theoretic metric with moderately wrong coefficients.

How productive would your company employees be if they could save two hours a day on regular tasks? With the growth and evolution of today's digital economy, companies face the challenge of managing increasingly complex business processes that involve massive amounts of data. This has also led to repetitive work, like requiring employees to manually perform data-intensive tasks when there are technologies available that could help free their time and automate tasks. According to a WorkMarket report, 53 percent of employees believe they could save up to two hours a day by automating tasks; that equates to roughly 20 hours a month. Working on tasks that could easily be automated is probably not the best use of employees' time, especially if your business is trying to improve productivity or customer service.

Action advising is a knowledge exchange mechanism between peers, namely student and teacher, that can help tackle exploration and sample inefficiency problems in deep reinforcement learning. Due to the practical limitations in peer-to-peer communication and the negative implications of over-advising, the peer responsible for initiating these interactions needs to do so only when it's most adequate to exchange advice. Most recently, student-initiated techniques that utilise state novelty and uncertainty estimations have obtained promising results. However, these estimations have several weaknesses, such as having no information regarding the characteristics of convergence and being subject to delays that occur in the presence of experience replay dynamics. We propose a student-initiated action advising algorithm that alleviates these shortcomings. Specifically, we employ Random Network Distillation (RND) to measure the novelty of an advice, for the student to determine whether to proceed with the request; furthermore, we perform RND updates only for the advised states to ensure that the student's convergence will not prevent it from utilising the teacher's knowledge at any stage of learning. Experiments in GridWorld and simplified versions of five Atari games show that our approach can perform on par with the state-of-the-art and demonstrate significant advantages in the scenarios where the existing methods are prone to fail.

For the next installment of the informal TechCrunch book club, we are reading the fourth story in Ted Chiang's Exhalation. The goal of this book club is to expand our minds to new worlds, ideas, and vistas, and The Lifecycle of Software Objects doesn't disappoint. Centered in a future world where virtual worlds and generalized AI have become commonplace, it's a fantastic example of speculative fiction that forces us to confront all kinds of fundamental questions. If you've missed the earlier parts in this book club series, be sure to check out: Some questions for the fifth story in the collection, Dacey's Patent Automatic Nanny, are included below. This is a much more sprawling story than the earlier short stories in Exhalation, with much more of a linear plot than the fractal koans we experienced before.

Like the city that hosts the Consumer Electronics Show (CES) there is a lot of noise on the show floor. Sifting through the lights, sounds and people can be an arduous task even for the most experienced CES attendees. Hidden past the North Hall of the Las Vegas Convention Center (LVCC) is a walkway to a tech oasis housed in the Westgate Hotel. This new area hosting SmartCity/IoT innovations is reminiscent of the old Eureka Park complete with folding tables and ballroom carpeting. The fact that such enterprises require their own area separate from the main halls of the LVCC and the startup pavilions of the Sands Hotel is an indication of how urbanization is being redefined by artificial intelligence.

We present a theoretical and experimental investigation of the quantization problem for artificial neural networks. We provide a mathematical definition of quantized neural networks and analyze their approximation capabilities, showing in particular that any Lipschitz-continuous map defined on a hypercube can be uniformly approximated by a quantized neural network. We then focus on the regularization effect of additive noise on the arguments of multi-step functions inherent to the quantization of continuous variables. In particular, when the expectation operator is applied to a non-differentiable multi-step random function, and if the underlying probability density is differentiable (in either classical or weak sense), then a differentiable function is retrieved, with explicit bounds on its Lipschitz constant. Based on these results, we propose a novel gradient-based training algorithm for quantized neural networks that generalizes the straight-through estimator, acting on noise applied to the network's parameters. We evaluate our algorithm on the CIFAR-10 and ImageNet image classification benchmarks, showing state-of-the-art performance on AlexNet and MobileNetV2 for ternary networks.

When we see demonstrations of robots planning for and performing multistep tasks, "it's almost always the case that a programmer has explicitly told the robot how to think about the world in order for it to make a plan," Konidaris said. "But if we want robots that can act more autonomously, they're going to need the ability to learn abstractions on their own." In computer science terms, these kinds of abstractions fall into two categories: "procedural abstractions" and "perceptual abstractions." Procedural abstractions are programs made out of low-level movements composed into higher-level skills. An example would be bundling all the little movements needed to open a door -- all the motor movements involved in reaching for the knob, turning it and pulling the door open -- into a single "open the door" skill. Once such a skill is built, you don't need to worry about how it works. All you need to know is when to run it. Roboticists -- including Konidaris himself -- have been studying how to make robots learn procedural abstractions for years, he says. But according to Konidaris, there's been less progress in perceptual abstraction, which has to do with helping a robot make sense of its pixelated surroundings.

When we see demonstrations of robots planning for and performing multistep tasks, "it's almost always the case that a programmer has explicitly told the robot how to think about the world in order for it to make a plan," Konidaris said. "But if we want robots that can act more autonomously, they're going to need the ability to learn abstractions on their own." In computer science terms, these kinds of abstractions fall into two categories: "procedural abstractions" and "perceptual abstractions." Procedural abstractions are programs made out of low-level movements composed into higher-level skills. An example would be bundling all the little movements needed to open a door--all the motor movements involved in reaching for the knob, turning it and pulling the door open--into a single "open the door" skill. Once such a skill is built, you don't need to worry about how it works. All you need to know is when to run it. Roboticists--including Konidaris himself--have been studying how to make robots learn procedural abstractions for years, he says. But according to Konidaris, there's been less progress in perceptual abstraction, which has to do with helping a robot make sense of its pixelated surroundings.