Pablo Barros

Active object tracking dynamically adjusts a camera’s field of view to maintain target visibility, unlike passive tracking where targets remain continuously observable. While substantial progress has been made in single-camera active tracking and passive multi-camera systems with overlapping views, many real-world camera networks are sequential and non-overlapping, forcing targets to traverse blind regions during transitions. In such settings, continuous tracking requires predictive coordination, explicit handoff reasoning, and control under partial observability challenges that remain largely unaddressed in existing work. This paper introduces NOVA-STAR, a system-level reinforcement learning framework for active tracking across sequential, non-overlapping multi-camera networks. NOVA-STAR employs Proximal Policy Optimization within a centralized control architecture and integrates a finite-state machine abstraction that explicitly encodes target visibility and handoff phases. Rather than replacing learning with fixed rules, the FSM provides structured temporal context that constrains the decision space, improves sample efficiency, and stabilizes policy learning during periods of target absence. Centralized observation sharing enables coordinated decision-making across cameras without the complexity of multi-agent learning, supporting practical deployment in controlled camera networks. To enable learning in this setting, a dedicated perception-independent simulation environment is developed that captures camera motion, visual discontinuities, and delayed target reappearance conditions that cannot be addressed using static datasets. Experimental results show that NOVA-STAR achieves 84.72% coordinated tracking, 82.32% handoff success, 86.92% reacquisition, and 97.50% standby accuracy. Ablation and behavioral analyses demonstrate the necessity of both the FSM abstraction and centralized coordination for reliable long-horizon tracking. A hybrid deployment with one physical and two simulated cameras further illustrates transferability beyond simulation. Overall, NOVA-STAR advances active multi-camera tracking by demonstrating that reinforcement learning, combined with structured state representations and deployable system design, can enable reliable coordination across non-overlapping camera networks.

In the social media age, platforms like X (formerly Twitter) are essential for spreading information and shaping public opinion. This initial study proposes a methodology to investigate interactions with Italian opinion leaders, forming clusters of leaders based on user engagement patterns while preserving user anonymity by analyzing the vocabulary eccentricity and sentiment analysis. Our findings reveal that positive sentiment is linked to a more eccentric vocabulary, particularly within specific clusters, indicating that sentiment is deeply linked to the community's shared language. Despite the prevalence of negative sentiment, most clusters feature a minority of leaders with eccentric, positive language. These consistent patterns over time suggest a potential universal behaviour for further study.

Reinforcement learning (RL)-based agents have demonstrated remarkable performance in multiplayer card game environments such as Chef's Hat. However, understanding why these agents excel in such dynamic and competitive settings remains a challenging endeavor. In this paper, we propose a novel method, named the “Mullet's Gambit“ to elucidate the strategies employed by RL-based agents within the context of the Chef's Hat card game. This method aims to provide insights into how RL-based agents navigate the complexities of multiplayer dynamics and assess their impact on opponents. By employing Mullet's Gambit, this investigation reveals the unique traits and efficacy of RL-based strategies compared to heuristic methodologies. This leads to the inference that RL-based agents not only acquire the skills to win but also to disrupt their opponents, thereby minimizing their potential actions.

In a competitive game scenario, a set of agents have to learn decisions that maximize their goals and minimize their adversaries' goals at the same time. Besides dealing with the increased dynamics of the scenarios due to the opponents' actions, they usually have to understand how to overcome the opponent's strategies. Most of the common solutions, usually based on continual learning or centralized multi-agent experiences, however, do not allow the development of personalized strategies to face individual opponents. In this paper, we propose a novel model composed of three neural layers that learn a representation of a competitive game, learn how to map the strategy of specific opponents, and how to disrupt them. The entire model is trained online, using a composed loss based on a contrastive optimization, to learn competitive and multiplayer games. We evaluate our model on a pokemon duel scenario and the four-player competitive Chef's Hat card game. Our experiments demonstrate that our model achieves better performance when playing against offline, online, and competitive-specific models, in particular when playing against the same opponent multiple times. We also present a discussion on the impact of our model, in particular on how well it deals with on specific strategy learning for each of the two scenarios.

Recent advances in reinforcement learning with social agents have allowed such models to achieve human-level performance on certain interaction tasks. However, most interactive scenarios do not have performance alone as an end-goal; instead, the social impact of these agents when interacting with humans is as important and largely unexplored. In this regard, this work proposes a novel reinforcement learning mechanism based on the social impact of rivalry behavior. Our proposed model aggregates objective and social perception mechanisms to derive a rivalry score that is used to modulate the learning of artificial agents. To investigate our proposed model, we design an interactive game scenario, using the Chef’s Hat Card Game, and examine how the rivalry modulation changes the agent’s playing style, and how this impacts the experience of human players on the game. Our results show that humans can detect specific social characteristics when playing against rival agents when compared to common agents, which affects directly the performance of the human players in subsequent games. We conclude our work by discussing how the different social and objective features that compose the artificial rivalry score contribute to our results.

Facial expression recognition, as part of an affective computing system, usually relies on solid performance metrics to be successful. These metrics depend significantly on the affective context in which one evaluates this system. While presenting excellent performance on the dataset it was trained on, a facial expression recognition model might drastically fail when one assesses it in a different scenario. Such performance reduction occurs because most facial perception models rely on an extreme generalization concept, focusing on a universal emotion perception system. With the recent findings on the non-universality of emotional perception, generalization of facial encoders seems not to be the optimal path to take. Therefore, exploiting transfer learning toward adapting specific facial features to specific scenarios could address this problem. This paper proposes and investigates a Spatial Transformer Plugin (STN) to rearrange different facial encoders towards particular affective representations from different scenarios. We experiment with our model in eight different facial expression recognition datasets (AffectNet and the derived MaskedAffectNet, OMG-Emotion, FERPlus, ElderReact, EmoReact, FABO and JAFFE datasets) and obtain competitive performance with much less training effort than state-of-the-art models. Besides performance alone, we introduce the STN as a mechanism towards a non-universal emotional perception system and discuss how it rearranges learned perception features to address some specific characteristics of each investigated dataset.

Current facial expression recognition systems demand an expensive re-training routine when deployed to different scenarios than they were trained for. Biasing them towards learning specific facial characteristics, instead of performing typical transfer learning methods, might help these systems to maintain high performance in different tasks, but with a reduced training effort. In this paper, we propose Contrastive Inhibitory Adaptati On (CIAO), a mechanism that adapts the last layer of facial encoders to depict specific affective characteristics on different datasets. CIAO presents an improvement in facial expression recognition performance over six different datasets with very unique affective representations, in particular when compared with state-of-the-art models. In our discussions, we make an in-depth analysis of how the learned high-level facial features are represented, and how they contribute to each individual dataset's characteristics. We finalize our study by discussing how CIAO positions itself within the range of recent findings on non-universal facial expressions perception, and its impact on facial expression recognition research.