Human beings always engage in a vast range of activities and tasks that demonstrate their ability to adapt to different scenarios. Any human activity can be represented as a temporal sequence of actions performed to achieve a certain goal. Unlike the time series datasets extracted from electronics or machines, these action sequences are highly disparate in their nature -- the time to finish a sequence of actions can vary between different persons. Therefore, understanding the dynamics of these sequences is essential for many downstream tasks such as activity length prediction, goal prediction, next action recommendation, etc. Existing neural network-based approaches that learn a continuous-time activity sequence (or CTAS) are limited to the presence of only visual data or are designed specifically for a particular task, i.e., limited to next action or goal prediction. In this paper, we present ProActive, a neural marked temporal point process (MTPP) framework for modeling the continuous-time distribution of actions in an activity sequence while simultaneously addressing three high-impact problems -- next action prediction, sequence-goal prediction, and end-to-end sequence generation. Specifically, we utilize a self-attention module with temporal normalizing flows to model the influence and the inter-arrival times between actions in a sequence. In addition, we propose a novel addition over the ProActive model that can handle variations in the order of actions, i.e., different methods of achieving a given goal. We demonstrate that this variant can learn the order in which the person or actor prefers to do their actions. Extensive experiments on sequences derived from three activity recognition datasets show the significant accuracy boost of ProActive over the state-of-the-art in terms of action and goal prediction, and the first-ever application of end-to-end action sequence generation.

Humans have always welcomed other beings in finance: over twenty years ago, some of the best Wall Street traders were outsmarted by Raven, a chimpanzee who picked stocks by throwing darts. Her index, called MonkeyDex, became one of the biggest sensations at the turn of the century after delivering a 213% gain. Perhaps because animals are not so easy to fit in offices, people have turned to other kinds of brains to choose equities. Big institutions are resorting to artificial intelligence (AI) to analyse stocks collating all sorts of information coming from a plethora of sources. In fact, while investments could previously be assessed based on financial reports and share price movement – what is called structured data – markets have been heavily influenced by unstructured data over the past few years.

Complex environments provide structured yet variable sensory inputs. To best exploit information from these environments, organisms must evolve the ability to correctly anticipate consequences of unknown stimuli, and act on these predictions. We propose an evolutionary path for neural networks, leading an organism from reactive behavior to simple proactive behavior and from simple proactive behavior to induction-based behavior. Through in-vitro and in-silico experiments, we define the minimal conditions necessary in a network with spike-timing dependent plasticity for the organism to go from reactive to proactive behavior. Our results support the existence of small evolutionary steps and four necessary conditions allowing embodied neural networks to evolve predictive and inductive abilities from an initial reactive strategy. We extend these conditions to more general structures.

Fifteen years ago, when the time became ripe for post-PC devices that put a premium on integrating software and hardware, Apple was the best-positioned company to lead the charge -- and it did. The company's vertical integration, its attention to detail and innovation in both software and hardware and its willingness to make big bets gave it an edge. And it used that edge to reel off its now-familiar string of game-changing products like the iPod, the iPhone, the MacBook Air and the iPad. Now, the iPod is essentially gone, and the other products are in mature or maturing markets, with either pretty flat or dropping sales. And the tech industry is turning to a new battlefield: Artificial intelligence, spread across many devices.

In this article we address the problem of realizing a service-providing reasoning infrastructure for pro-active humanassistance in intelligent environments. We propose SAM, an architecture which leverages temporal knowledge represented asrelations in Allen’s interval algebra and constraint-based temporal planning techniques. SAM provides two key capabilities forcontextualized service provision: human activity recognition and planning for controlling pervasive actuation devices. Whiledrawing inspiration from several state-of-the-art approaches, SAM provides a unique feature which has thus far not been addressed in the literature, namely the seamless integration of these two key capabilities. It does so by leveraging a constraint-basedreasoning paradigm whereby both requirements for recognition and for planning/execution are represented as constraints andreasoned upon continuously.

In this article we address the problem of realizing a service-providing reasoning infrastructure for pro-active humanassistance in intelligent environments. We propose SAM, an architecture which leverages temporal knowledge represented asrelations in Allen’s interval algebra and constraint-based temporal planning techniques. SAM provides two key capabilities forcontextualized service provision: human activity recognition and planning for controlling pervasive actuation devices. Whiledrawing inspiration from several state-of-the-art approaches, SAM provides a unique feature which has thus far not been addressed in the literature, namely the seamless integration of these two key capabilities. It does so by leveraging a constraint-basedreasoning paradigm whereby both requirements for recognition and for planning/execution are represented as constraints andreasoned upon continuously.

In this article we address the problem of realizing a service-providing reasoning infrastructure for pro-active humanassistance in intelligent environments. We propose SAM, an architecture which leverages temporal knowledge represented asrelations in Allen’s interval algebra and constraint-based temporal planning techniques. SAM provides two key capabilities forcontextualized service provision: human activity recognition and planning for controlling pervasive actuation devices. Whiledrawing inspiration from several state-of-the-art approaches, SAM provides a unique feature which has thus far not been addressed in the literature, namely the seamless integration of these two key capabilities. It does so by leveraging a constraint-basedreasoning paradigm whereby both requirements for recognition and for planning/execution are represented as constraints andreasoned upon continuously.

In this article we address the problem of realizing a service-providing reasoning infrastructure for pro-active humanassistance in intelligent environments. We propose SAM, an architecture which leverages temporal knowledge represented asrelations in Allen’s interval algebra and constraint-based temporal planning techniques. SAM provides two key capabilities forcontextualized service provision: human activity recognition and planning for controlling pervasive actuation devices. Whiledrawing inspiration from several state-of-the-art approaches, SAM provides a unique feature which has thus far not been addressed in the literature, namely the seamless integration of these two key capabilities. It does so by leveraging a constraint-basedreasoning paradigm whereby both requirements for recognition and for planning/execution are represented as constraints andreasoned upon continuously.

In this article we address the problem of realizing a service-providing reasoning infrastructure for pro-active humanassistance in intelligent environments. We propose SAM, an architecture which leverages temporal knowledge represented asrelations in Allen’s interval algebra and constraint-based temporal planning techniques. SAM provides two key capabilities forcontextualized service provision: human activity recognition and planning for controlling pervasive actuation devices. Whiledrawing inspiration from several state-of-the-art approaches, SAM provides a unique feature which has thus far not been addressed in the literature, namely the seamless integration of these two key capabilities. It does so by leveraging a constraint-basedreasoning paradigm whereby both requirements for recognition and for planning/execution are represented as constraints andreasoned upon continuously.

In this article we address the problem of realizing a service-providing reasoning infrastructure for pro-active humanassistance in intelligent environments. We propose SAM, an architecture which leverages temporal knowledge represented asrelations in Allen’s interval algebra and constraint-based temporal planning techniques. SAM provides two key capabilities forcontextualized service provision: human activity recognition and planning for controlling pervasive actuation devices. Whiledrawing inspiration from several state-of-the-art approaches, SAM provides a unique feature which has thus far not been addressed in the literature, namely the seamless integration of these two key capabilities. It does so by leveraging a constraint-basedreasoning paradigm whereby both requirements for recognition and for planning/execution are represented as constraints andreasoned upon continuously.