Finland is the first country having released its AI strategy in Europe already in March 2017. According to a study committed by Accenture and Frontier Economics, Finland ranked second that year, after the US, among the 11 developed countries in which economic growth potential is made possible by AI. According to Finland, this is because of the country's business structure (technologically intensive) and the public sector degree of digitalisation (see Finland, 2017, p. 12). The national strategy has been commissioned by the Government of Juha Sipilä to the Ministry of Economic Affairs and Employment, which in turn has nominated a steering group on AI to work on the national strategy. The AI Working Group has released the first draft of the strategy in 2017, though the work on the optimum public policies to be implemented is actually an on-going process, which has already been updated in 2019.

Advancements in machine learning have recently enabled the hyper-realistic synthesis of prose, images, audio and video data, in what is referred to as artificial intelligence (AI)-generated media. These techniques offer novel opportunities for creating interactions with digital portrayals of individuals that can inspire and intrigue us. AI-generated portrayals of characters can feature synthesized faces, bodies and voices of anyone, from a fictional character to a historical figure, or even a deceased family member. Although negative use cases of this technology have dominated the conversation so far, in this Perspective we highlight emerging positive use cases of AI-generated characters, specifically in supporting learning and well-being. We demonstrate an easy-to-use AI character generation pipeline to enable such outcomes and discuss ethical implications as well as the need for including traceability to help maintain trust in the generated media.

Lance Fortnow (lfortnow@iit.edu) is a professor and dean of the College of Computing at Illinois Institute of Technology, Chicago, IL, USA.

It should no longer be a surprise that algorithms can discriminate. A criminal risk-assessment algorithm is far more likely to erroneously predict a Black defendant will commit a crime in the future than a white defendant.2 Ad-targeting algorithms promote job opportunities to race- and gender-skewed audiences, showing secretary and supermarket job ads to far more women than men.1 A hospital's resource-allocation algorithm favored white over Black patients with the same level of medical need.5 Algorithmic discrimination is particularly troubling when it affects consequential social decisions, such as who gets released from jail, or has access to a loan or health care. Employment is a prime example. Employers are increasingly relying on algorithmic tools to recruit, screen, and select job applicants by making predictions about which candidates will be good employees.

Artificial intelligence is everywhere these days. The National AI Initiative Act became law in the U.S. on Jan. 1, 2021, aiming "to accelerate AI research and application for the Nation's economic prosperity and national security." The U.S. National Science Foundation launched in 2020 several AI Research Institutes to push forward the frontiers of artificial intelligence. One of the themes of this research initiative is "AI-Augmented Learning." This quest to improve education via technology reminds me of "Profession;" a 1957 science-fiction story by Isaac Asimov.

Recently, barrier function-based safe reinforcement learning (RL) with the actor-critic structure for continuous control tasks has received increasing attention. It is still challenging to learn a near-optimal control policy with safety and convergence guarantees. Also, few works have addressed the safe RL algorithm design under time-varying safety constraints. This paper proposes a model-based safe RL algorithm for optimal control of nonlinear systems with time-varying state and control constraints. In the proposed approach, we construct a novel barrier-based control policy structure that can guarantee control safety. A multi-step policy evaluation mechanism is proposed to predict the policy's safety risk under time-varying safety constraints and guide the policy to update safely. Theoretical results on stability and robustness are proven. Also, the convergence of the actor-critic learning algorithm is analyzed. The performance of the proposed algorithm outperforms several state-of-the-art RL algorithms in the simulated Safety Gym environment. Furthermore, the approach is applied to the integrated path following and collision avoidance problem for two real-world intelligent vehicles. A differential-drive vehicle and an Ackermann-drive one are used to verify the offline deployment performance and the online learning performance, respectively. Our approach shows an impressive sim-to-real transfer capability and a satisfactory online control performance in the experiment.

The performance of a distillation-based compressed network is governed by the quality of distillation. The reason for the suboptimal distillation of a large network (teacher) to a smaller network (student) is largely attributed to the gap in the learning capacities of given teacher-student pair. While it is hard to distill all the knowledge of a teacher, the quality of distillation can be controlled to a large extent to achieve better performance. Our experiments show that the quality of distillation is largely governed by the quality of teacher's response, which in turn is heavily affected by the presence of similarity information in its response. A well-trained large capacity teacher loses similarity information between classes in the process of learning fine-grained discriminative properties for classification. The absence of similarity information causes the distillation process to be reduced from one example-many class learning to one example-one class learning, thereby throttling the flow of diverse knowledge from the teacher. With the implicit assumption that only the instilled knowledge can be distilled, instead of focusing only on the knowledge distilling process, we scrutinize the knowledge inculcation process. We argue that for a given teacher-student pair, the quality of distillation can be improved by finding the sweet spot between batch size and number of epochs while training the teacher. We discuss the steps to find this sweet spot for better distillation. We also propose the distillation hypothesis to differentiate the behavior of the distillation process between knowledge distillation and regularization effect. We conduct all our experiments on three different datasets.

This paper studies continual learning (CL) for sentiment classification (SC). In this setting, the CL system learns a sequence of SC tasks incrementally in a neural network, where each task builds a classifier to classify the sentiment of reviews of a particular product category or domain. Two natural questions are: Can the system transfer the knowledge learned in the past from the previous tasks to the new task to help it learn a better model for the new task? And, can old models for previous tasks be improved in the process as well? This paper proposes a novel technique called KAN to achieve these objectives. KAN can markedly improve the SC accuracy of both the new task and the old tasks via forward and backward knowledge transfer. The effectiveness of KAN is demonstrated through extensive experiments.

Existing research on continual learning of a sequence of tasks focused on dealing with catastrophic forgetting, where the tasks are assumed to be dissimilar and have little shared knowledge. Some work has also been done to transfer previously learned knowledge to the new task when the tasks are similar and have shared knowledge. To the best of our knowledge, no technique has been proposed to learn a sequence of mixed similar and dissimilar tasks that can deal with forgetting and also transfer knowledge forward and backward. This paper proposes such a technique to learn both types of tasks in the same network. For dissimilar tasks, the algorithm focuses on dealing with forgetting, and for similar tasks, the algorithm focuses on selectively transferring the knowledge learned from some similar previous tasks to improve the new task learning. Additionally, the algorithm automatically detects whether a new task is similar to any previous tasks. Empirical evaluation using sequences of mixed tasks demonstrates the effectiveness of the proposed model.

We present a new neighbor sampling method on temporal graphs. In a temporal graph, predicting different nodes' time-varying properties can require the receptive neighborhood of various temporal scales. In this work, we propose the TNS (Time-aware Neighbor Sampling) method: TNS learns from temporal information to provide an adaptive receptive neighborhood for every node at any time. Learning how to sample neighbors is non-trivial, since the neighbor indices in time order are discrete and not differentiable. To address this challenge, we transform neighbor indices from discrete values to continuous ones by interpolating the neighbors' messages. TNS can be flexibly incorporated into popular temporal graph networks to improve their effectiveness without increasing their time complexity. TNS can be trained in an end-to-end manner. It needs no extra supervision and is automatically and implicitly guided to sample the neighbors that are most beneficial for prediction. Empirical results on multiple standard datasets show that TNS yields significant gains on edge prediction and node classification.