Current trends in networking propose the use of Machine Learning (ML) for a wide variety of network optimization tasks. As such, many efforts have been made to produce ML-based solutions for Traffic Engineering (TE), which is a fundamental problem in ISP networks. Nowadays, state-of-the-art TE optimizers rely on traditional optimization techniques, such as Local search, Constraint Programming, or Linear programming. In this paper, we present MAGNNETO, a distributed ML-based framework that leverages Multi-Agent Reinforcement Learning and Graph Neural Networks for distributed TE optimization. MAGNNETO deploys a set of agents across the network that learn and communicate in a distributed fashion via message exchanges between neighboring agents. Particularly, we apply this framework to optimize link weights in OSPF, with the goal of minimizing network congestion. In our evaluation, we compare MAGNNETO against several state-of-the-art TE optimizers in more than 75 topologies (up to 153 nodes and 354 links), including realistic traffic loads. Our experimental results show that, thanks to its distributed nature, MAGNNETO achieves comparable performance to state-of-the-art TE optimizers with significantly lower execution times. Moreover, our ML-based solution demonstrates a strong generalization capability to successfully operate in new networks unseen during training.

In recent years, imitation learning (IL) has been widely used in industry as the core of autonomous vehicle (AV) planning modules. However, previous IL works show sample inefficiency and low generalisation in safety-critical scenarios, on which they are rarely tested. As a result, IL planners can reach a performance plateau where adding more training data ceases to improve the learnt policy. First, our work presents an IL model using the spline coefficient parameterisation and offline expert queries to enhance safety and training efficiency. Then, we expose the weakness of the learnt IL policy by synthetically generating critical scenarios through optimisation of parameters of the driver's risk field (DRF), a parametric human driving behaviour model implemented in a multi-agent traffic simulator based on the Lyft Prediction Dataset. To continuously improve the learnt policy, we retrain the IL model with augmented data. Thanks to the expressivity and interpretability of the DRF, the desired driving behaviours can be encoded and aggregated to the original training data. Our work constitutes a full development cycle that can efficiently and continuously improve the learnt IL policies in closed-loop. Finally, we show that our IL planner developed with less training resource still has superior performance compared to the previous state-of-the-art.

In this paper we present a grammar and control synthesis framework for online modification of Event-based Signal Temporal Logic (STL) specifications, during execution. These modifications allow a user to change the robots' task in response to potential future violations, changes to the environment, or user-defined task design changes. In cases where a modification is not possible, we provide feedback to the user and suggest alternative modifications. We demonstrate our task modification process using a Hello Robot Stretch satisfying an Event-based STL specification.

'Reincarnation' in reinforcement learning has been proposed as a formalisation of reusing prior computation from past experiments when training an agent in an environment. In this paper, we present a brief foray into the paradigm of reincarnation in the multi-agent (MA) context. We consider the case where only some agents are reincarnated, whereas the others are trained from scratch -- selective reincarnation. In the fully-cooperative MA setting with heterogeneous agents, we demonstrate that selective reincarnation can lead to higher returns than training fully from scratch, and faster convergence than training with full reincarnation. However, the choice of which agents to reincarnate in a heterogeneous system is vitally important to the outcome of the training -- in fact, a poor choice can lead to considerably worse results than the alternatives. We argue that a rich field of work exists here, and we hope that our effort catalyses further energy in bringing the topic of reincarnation to the multi-agent realm.

Object-centric process discovery (OCPD) constitutes a paradigm shift in process mining. Instead of assuming a single case notion present in the event log, OCPD can handle events without a single case notion, but that are instead related to a collection of objects each having a certain type. The object types constitute multiple, interacting case notions. The output of OCPD is an object-centric Petri net, i.e. a Petri net with object-typed places, that represents the parallel execution of multiple execution flows corresponding to object types. Similar to classical process discovery, where we aim for behaviorally sound process models as a result, in OCPD, we aim for soundness of the resulting object-centric Petri nets. However, the existing OCPD approach can result in violations of soundness. As we will show, one violation arises for multiple interacting object types with loops that arise in collaborative systems. This paper proposes an extended OCPD approach and proves that it does not suffer from this violation of soundness of the resulting object-centric Petri nets. We also show how we prevent the OCPD approach from introducing spurious interactions in the discovered object-centric Petri net. The proposed framework is prototypically implemented.

Research in model-based reinforcement learning has made significant progress in recent years. Compared to single-agent settings, the exponential dimension growth of the joint state-action space in multi-agent systems dramatically increases the complexity of the environment dynamics, which makes it infeasible to learn an accurate global model and thus necessitates the use of agent-wise local models. However, during multi-step model rollouts, the prediction of one local model can affect the predictions of other local models in the next step. As a result, local prediction errors can be propagated to other localities and eventually give rise to considerably large global errors. Furthermore, since the models are generally used to predict for multiple steps, simply minimizing one-step prediction errors regardless of their long-term effect on other models may further aggravate the propagation of local errors. To this end, we propose Models as AGents (MAG), a multi-agent model optimization framework that reversely treats the local models as multi-step decision making agents and the current policies as the dynamics during the model rollout process. In this way, the local models are able to consider the multi-step mutual affect between each other before making predictions. Theoretically, we show that the objective of MAG is approximately equivalent to maximizing a lower bound of the true environment return. Experiments on the challenging StarCraft II benchmark demonstrate the effectiveness of MAG.

Object goal navigation (ObjectNav) in unseen environments is a fundamental task for Embodied AI. Agents in existing works learn ObjectNav policies based on 2D maps, scene graphs, or image sequences. Considering this task happens in 3D space, a 3D-aware agent can advance its ObjectNav capability via learning from fine-grained spatial information. However, leveraging 3D scene representation can be prohibitively unpractical for policy learning in this floor-level task, due to low sample efficiency and expensive computational cost. In this work, we propose a framework for the challenging 3D-aware ObjectNav based on two straightforward sub-policies. The two sub-polices, namely corner-guided exploration policy and category-aware identification policy, simultaneously perform by utilizing online fused 3D points as observation. Through extensive experiments, we show that this framework can dramatically improve the performance in ObjectNav through learning from 3D scene representation. Our framework achieves the best performance among all modular-based methods on the Matterport3D and Gibson datasets, while requiring (up to 30x) less computational cost for training.

Multi-Agent Pickup and Delivery (MAPD) is the problem of computing collision-free paths for a group of agents such that they can safely reach delivery locations from pickup ones. These locations are provided at runtime, making MAPD a combination between classical Multi-Agent Path Finding (MAPF) and online task assignment. Current algorithms for MAPD do not consider many of the practical issues encountered in real applications: real agents often do not follow the planned paths perfectly, and may be subject to delays and failures. In this paper, we study the problem of MAPD with delays, and we present two solution approaches that provide robustness guarantees by planning paths that limit the effects of imperfect execution. In particular, we introduce two algorithms, k-TP and p-TP, both based on a decentralized algorithm typically used to solve MAPD, Token Passing (TP), which offer deterministic and probabilistic guarantees, respectively. Experimentally, we compare our algorithms against a version of TP enriched with online replanning. k-TP and p-TP provide robust solutions, significantly reducing the number of replans caused by delays, with little or no increase in solution cost and running time.

Episodic training, where an agent's environment is reset after every success or failure, is the de facto standard when training embodied reinforcement learning (RL) agents. The underlying assumption that the environment can be easily reset is limiting both practically, as resets generally require human effort in the real world and can be computationally expensive in simulation, and philosophically, as we'd expect intelligent agents to be able to continuously learn without intervention. Work in learning without any resets, i.e{.} Reset-Free RL (RF-RL), is promising but is plagued by the problem of irreversible transitions (e.g{.} an object breaking) which halt learning. Moreover, the limited state diversity and instrument setup encountered during RF-RL means that works studying RF-RL largely do not require their models to generalize to new environments. In this work, we instead look to minimize, rather than completely eliminate, resets while building visual agents that can meaningfully generalize. As studying generalization has previously not been a focus of benchmarks designed for RF-RL, we propose a new Stretch Pick-and-Place benchmark designed for evaluating generalizations across goals, cosmetic variations, and structural changes. Moreover, towards building performant reset-minimizing RL agents, we propose unsupervised metrics to detect irreversible transitions and a single-policy training mechanism to enable generalization. Our proposed approach significantly outperforms prior episodic, reset-free, and reset-minimizing approaches achieving higher success rates with fewer resets in Stretch-P\&P and another popular RF-RL benchmark. Finally, we find that our proposed approach can dramatically reduce the number of resets required for training other embodied tasks, in particular for RoboTHOR ObjectNav we obtain higher success rates than episodic approaches using 99.97\% fewer resets.

The current state of the art systems in Artificial Intelligence (AI) enabled intrusion detection use a variety of black box methods. These black box methods are generally trained using Error Based Learning (EBL) techniques with a focus on creating accurate models. These models have high performative costs and are not easily explainable. A white box Competitive Learning (CL) based eXplainable Intrusion Detection System (X-IDS) offers a potential solution to these problem. CL models utilize an entirely different learning paradigm than EBL approaches. This different learning process makes the CL family of algorithms innately explainable and less resource intensive. In this paper, we create an X-IDS architecture that is based on DARPA's recommendation for explainable systems. In our architecture we leverage CL algorithms like, Self Organizing Maps (SOM), Growing Self Organizing Maps (GSOM), and Growing Hierarchical Self Organizing Map (GHSOM). The resulting models can be data-mined to create statistical and visual explanations. Our architecture is tested using NSL-KDD and CIC-IDS-2017 benchmark datasets, and produces accuracies that are 1% - 3% less than EBL models. However, CL models are much more explainable than EBL models. Additionally, we use a pruning process that is able to significantly reduce the size of these CL based models. By pruning our models, we are able to increase prediction speeds. Lastly, we analyze the statistical and visual explanations generated by our architecture, and we give a strategy that users could use to help navigate the set of explanations. These explanations will help users build trust with an Intrusion Detection System (IDS), and allow users to discover ways to increase the IDS's potency.