Time series classification (TSC) is a challenging task that attracted many researchers in the last few years. One main challenge in TSC is the diversity of domains where time series data come from. Thus, there is no "one model that fits all" in TSC. Some algorithms are very accurate in classifying a specific type of time series when the whole series is considered, while some only target the existence/non-existence of specific patterns/shapelets. Yet other techniques focus on the frequency of occurrences of discriminating patterns/features. This paper presents a new classification technique that addresses the inherent diversity problem in TSC using a nature-inspired method. The technique is stimulated by how flies look at the world through "compound eyes" that are made up of thousands of lenses, called ommatidia. Each ommatidium is an eye with its own lens, and thousands of them together create a broad field of vision. The developed technique similarly uses different lenses and representations to look at the time series, and then combines them for broader visibility. These lenses have been created through hyper-parameterisation of symbolic representations (Piecewise Aggregate and Fourier approximations). The algorithm builds a random forest for each lens, then performs soft dynamic voting for classifying new instances using the most confident eyes, i.e, forests. We evaluate the new technique, coined Co-eye, using the recently released extended version of UCR archive, containing more than 100 datasets across a wide range of domains. The results show the benefits of bringing together different perspectives reflecting on the accuracy and robustness of Co-eye in comparison to other state-of-the-art techniques.

Despite success in many real-world tasks (e.g., robotics), reinforcement learning (RL) agents still learn from tabula rasa when facing new and dynamic scenarios. By contrast, humans can offload this burden through textual descriptions. Although recent works have shown the benefits of instructive texts in goal-conditioned RL, few have studied whether descriptive texts help agents to generalize across dynamic environments. To promote research in this direction, we introduce a new platform, BabyAI++, to generate various dynamic environments along with corresponding descriptive texts. Moreover, we benchmark several baselines inherited from the instruction following setting and develop a novel approach towards visually-grounded language learning on our platform. Extensive experiments show strong evidence that using descriptive texts improves the generalization of RL agents across environments with varied dynamics.

In this report, we will define a new approach to the problem of non deterministic planning for extended temporal goals. In particular, we will give a solution to this problem reducing it to a fully observable non deterministic (FOND) planning problem and taking advantage of the LTLfToDFA tool. First of all, we will introduce the main idea and motivations supporting our approach. Then, we will give some preliminaries explaining the Planning Domain Definition Language (PDDL) language and the FOND planning problem formally. After that, we will illustrate our FOND4LTLfPLTLf (also available online) approach with the encoding of temporal goals into a PDDL domain and problem. Finally, we will present some of the results obtained through the application of the proposed solution.

Summarizing video content is an important task in many applications. This task can be defined as the computation of the ordered list of actions present in a video. Such a list could be extracted using action detection algorithms. However, it is not necessary to determine the temporal boundaries of actions to know their existence. Moreover, localizing precise boundaries usually requires dense video analysis to be effective. In this work, we propose to directly compute this ordered list by sparsely browsing the video and selecting one frame per action instance, task known as action spotting in literature. To do this, we propose ActionSpotter, a spotting algorithm that takes advantage of Deep Reinforcement Learning to efficiently spot actions while adapting its video browsing speed, without additional supervision. Experiments performed on datasets THUMOS14 and ActivityNet show that our framework outperforms state of the art detection methods. In particular, the spotting mean Average Precision on THUMOS14 is significantly improved from 59.7% to 65.6% while skipping 23% of video.

Interpretability of machine learning models has gained more and more attention among researchers in the artificial intelligence (AI) and human-computer interaction (HCI) communities. Most existing work focuses on decision making, whereas we consider knowledge discovery. In particular, we focus on evaluating AI-discovered knowledge/rules in the arts and humanities. From a specific scenario, we present an experimental procedure to collect and assess human-generated verbal interpretations of AI-generated music theory/rules rendered as sophisticated symbolic/numeric objects. Our goal is to reveal both the possibilities and the challenges in such a process of decoding expressive messages from AI sources. We treat this as a first step towards 1) better design of AI representations that are human interpretable and 2) a general methodology to evaluate interpretability of AI-discovered knowledge representations.

ABSTRACT Mirror Ritual is an interactive installation that challenges the existing paradigms in our understanding of human emotion and machine perception. In contrast to prescriptive interfaces, the work's real-time affective interface engages the audience in the iterative conceptualisation of their emotional state through the use of affectively-charged machine generated poetry. The audience are encouraged to make sense of the mirror's poetry by framing it with respect to their recent life experiences, effectively'putting into words' their felt emotion. This process of affect labelling and contextualisation works to not only regulate emotion, but helps to construct the rich personal narratives that constitute human identity. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page.

Existing Visual Question Answering (VQA) methods tend to exploit dataset biases and spurious statistical correlations, instead of producing right answers for the right reasons. To address this issue, recent bias mitigation methods for VQA propose to incorporate visual cues (e.g., human attention maps) to better ground the VQA models, showcasing impressive gains. However, we show that the performance improvements are not a result of improved visual grounding, but a regularization effect which prevents over-fitting to linguistic priors. For instance, we find that it is not actually necessary to provide proper, human-based cues; random, insensible cues also result in similar improvements. Based on this observation, we propose a simpler regularization scheme that does not require any external annotations and yet achieves near state-of-the-art performance on VQA-CPv2.

Since its launch in 2019, 5G has proven far superior to the existing 4G network. Capable of transforming data faster and more efficiently with less latency and interference, 5G technology has the capacity to serve a greater number of connected devices and to have profound implications across almost every sector of society -- bringing automation to industries including construction, financial services and healthcare, to name just a few. One area we can expect to see transformed is the automotive and autonomous vehicle (AV) sector, as 5G enables automakers to create smarter vehicles that are better able to handle a vast array of new data and, as a result, a range of road conditions. BMW has already announced plans to launch the first premium vehicle to carry standard 5G wireless in 2021. Meanwhile, Volkswagen plans to begin constructing its own 5G mobile networks in Germany this year.

Artificial intelligence seems to be everywhere, but what we are really witnessing is a supervised-learning revolution: We teach computers to see patterns, much as we teach children to read. But the future of A.I. depends on computer systems that learn on their own, without supervision, researchers say. When a mother points to a dog and tells her baby, "Look at the doggy," the child learns what to call the furry four-legged friends. But when that baby stands and stumbles, again and again, until she can walk, that is something else. Just as humans learn mostly through observation or trial and error, computers will have to go beyond supervised learning to reach the holy grail of human-level intelligence.

Over the past few years, tidymodels has been gradually emerging as the tidyverse's machine learning toolkit. Well, it turns out that R has a consistency problem. Since everything was made by different people and using different principles, everything has a slightly different interface, and trying to keep everything in line can be frustrating. Several years ago, Max Kuhn (formerly at Pfeizer, now at RStudio) developed the caret R package (see my caret tutorial) aimed at creating a uniform interface for the massive variety of machine learning models that exist in R. Caret was great in a lot of ways, but also limited in others. In my own use, I found it to be quite slow whenever I tried to use on problems of any kind of modest size.