A successful tactic that is followed by the scientific community for advancing AI is to treat games as problems, which has been proven to lead to various breakthroughs. We adapt this strategy in order to study Rocket League, a widely popular but rather under-explored 3D multiplayer video game with a distinct physics engine and complex dynamics that pose a significant challenge in developing efficient and high-performance game-playing agents. In this paper, we present Lucy-SKG, a Reinforcement Learning-based model that learned how to play Rocket League in a sample-efficient manner, outperforming by a notable margin the two highest-ranking bots in this game, namely Necto (2022 bot champion) and its successor Nexto, thus becoming a state-of-the-art agent. Our contributions include: a) the development of a reward analysis and visualization library, b) novel parameterizable reward shape functions that capture the utility of complex reward types via our proposed Kinesthetic Reward Combination (KRC) technique, and c) design of auxiliary neural architectures for training on reward prediction and state representation tasks in an on-policy fashion for enhanced efficiency in learning speed and performance. By performing thorough ablation studies for each component of Lucy-SKG, we showed their independent effectiveness in overall performance. In doing so, we demonstrate the prospects and challenges of using sample-efficient Reinforcement Learning techniques for controlling complex dynamical systems under competitive team-based multiplayer conditions.

Language Models are the underpin of all modern Natural Language Processing (NLP) tasks. The introduction of the Transformers architecture has contributed significantly into making Language Modeling very effective across many NLP task, leading to significant advancements in the field. However, Transformers come with a big computational cost, which grows quadratically with respect to the input length. This presents a challenge as to understand long texts requires a lot of context. In this paper, we propose a Fine-Tuning framework, named CoreLM, that extends the architecture of current Pretrained Language Models so that they incorporate explicit entity information. By introducing entity representations, we make available information outside the contextual space of the model, which results in a better Language Model for a fraction of the computational cost. We implement our approach using GPT2 and compare the fine-tuned model to the original. Our proposed model achieves a lower Perplexity in GUMBY and LAMBDADA datasets when compared to GPT2 and a fine-tuned version of GPT2 without any changes. We also compare the models' performance in terms of Accuracy in LAMBADA and Children's Book Test, with and without the use of model-created coreference annotations.

Entity Coreference Resolution is the task of resolving all the mentions in a document that refer to the same real world entity and is considered as one of the most difficult tasks in natural language understanding. While in it is not an end task, it has been proved to improve downstream natural language processing tasks such as entity linking, machine translation, summarization and chatbots. We conducted a systematic a review of neural-based approached and provide a detailed appraisal of the datasets and evaluation metrics in the field. Emphasis is given on Pronoun Resolution, a subtask of Coreference Resolution, which has seen various improvements in the recent years. We conclude the study by highlight the lack of agreed upon standards and propose a way to expand the task even further.

Background: In this paper we present the approaches and methods employed in order to deal with a large scale multi-label semantic indexing task of biomedical papers. This work was mainly implemented within the context of the BioASQ challenge of 2014. Methods: The main contribution of this work is a multi-label ensemble method that incorporates a McNemar statistical significance test in order to validate the combination of the constituent machine learning algorithms. Some secondary contributions include a study on the temporal aspects of the BioASQ corpus (observations apply also to the BioASQ's super-set, the PubMed articles collection) and the proper adaptation of the algorithms used to deal with this challenging classification task. Results: The ensemble method we developed is compared to other approaches in experimental scenarios with subsets of the BioASQ corpus giving positive results. During the BioASQ 2014 challenge we obtained the first place during the first batch and the third in the two following batches. Our success in the BioASQ challenge proved that a fully automated machine-learning approach, which does not implement any heuristics and rule-based approaches, can be highly competitive and outperform other approaches in similar challenging contexts.

One of the most promising trends in Domain Independent AI Planning, nowadays, is state-space heuristic planning. The planners of this category construct general but efficient heuristic functions, which are used as a guide to traverse the state space either in a forward or a in backward direction. Although specific problems may favor one or the other direction, there is no clear evidence why any of them should be generally preferred. This paper proposes a hybrid search strategy that combines search in both directions. The search begins from the Initial State in a forward direction and proceeds with a weighted A* search until no further improving states can be found. At that point, the algorithm changes direction and starts regressing the Goals trying to reach the best state found at the previous step. The direction of the search may change several times before a solution can be found. Two domain-independent heuristic functions based on ASP/HSP planners enhanced with a Goal Ordering technique have been implemented. The whole bi-directional planning system, named BP, was tested on a variety of problems adopted from the recent AIPS-00 planning competition with quite promising results. The paper also discusses the subject of domain analysis for state-space planning and proposes two methods for the elimination of redundant information from the problem definition and for the identification of independent sub-problems.

When transferring knowledge between reinforcement learning agents with different state representations or actions, past knowledge must be efficiently mapped so that it assists learning. The majority of the existing approaches use pre-defined mappings given by a domain expert. To overcome this limitations and allow autonomous transfer learning, this paper introduces a method for weighting and using multiple inter-task mappings, named COMBREL. Experimental results show that the use of multiple inter-task mappings, accompanied with a selection mechanism, can significantly boost the performance of transfer learning, relative to learning without transfer and relative to using a single hand-picked mapping.

This article presents the GRT planner, a forward heuristic state-space planner, and comments on the results obtained from the Fifth International Conference on Artificial Intelligence Planning and Scheduling (AIPS'00) planning competition. The grt planner works in two phases. In the preprocessing phase, it estimates the distances between the facts and the goals of the problem. During the search phase, the estimates are used to guide a forward-directed search.

The main idea that arise during the forward search phase and of the planner is to compute offline, in the preprocessing the goals. This approach succeeds in the notion of related facts in the goal-regression avoiding computing estimates for invalid facts process. These are facts that have been achieved in the preprocessing phase. However, it introduces either by the same or subsequent actions, without some problems in situations where the the last actions deleting the facts achieved goal state is not completely described because first. The cost of achieving simultaneously a set an action to regress the goals might not exist. of unrelated facts is considered equal to the To cope with this situation, at the beginning sum of their individual costs, whereas the cost of the preprocessing phase, We know from our experience that if move actions were Table 1.