arXiv subjects

Michael Kölle

Publications and source records attributed to Michael Kölle.

At least 55 records · Page 3Linked to original sources

arXiv2024

Aquarium: A Comprehensive Framework for Exploring Predator-Prey Dynamics through Multi-Agent Reinforcement Learning Algorithms

Recent advances in Multi-Agent Reinforcement Learning have prompted the modeling of intricate interactions between agents in simulated environments. In particular, the predator-prey dynamics have captured substantial interest and various simulations been tailored to unique requirements. To prevent further time-intensive developments, we introduce Aquarium, a comprehensive Multi-Agent Reinforcement Learning environment for predator-prey interaction, enabling the study of emergent behavior. Aquarium is open source and offers a seamless integration of the PettingZoo framework, allowing a quick start with proven algorithm implementations. It features physics-based agent movement on a two-dimensional, edge-wrapping plane. The agent-environment interaction (observations, actions, rewards) and the environment settings (agent speed, prey reproduction, predator starvation, and others) are fully customizable. Besides a resource-efficient visualization, Aquarium supports to record video files, providing a visual comprehension of agent behavior. To demonstrate the environment's capabilities, we conduct preliminary studies which use PPO to train multiple prey agents to evade a predator. In accordance to the literature, we find Individual Learning to result in worse performance than Parameter Sharing, which significantly improves coordination and sample-efficiency.

cs.AI↗

arXiv2024

Introducing Reduced-Width QNNs, an AI-inspired Ansatz Design Pattern

Variational Quantum Algorithms are one of the most promising candidates to yield the first industrially relevant quantum advantage. Being capable of arbitrary function approximation, they are often referred to as Quantum Neural Networks (QNNs) when being used in analog settings as classical Artificial Neural Networks (ANNs). Similar to the early stages of classical machine learning, known schemes for efficient architectures of these networks are scarce. Exploring beyond existing design patterns, we propose a reduced-width circuit ansatz design, which is motivated by recent results gained in the analysis of dropout regularization in QNNs. More precisely, this exploits the insight, that the gates of overparameterized QNNs can be pruned substantially until their expressibility decreases. The results of our case study show, that the proposed design pattern can significantly reduce training time while maintaining the same result quality as the standard "full-width" design in the presence of noise.

quant-ph↗

arXiv2024

ClusterComm: Discrete Communication in Decentralized MARL using Internal Representation Clustering

In the realm of Multi-Agent Reinforcement Learning (MARL), prevailing approaches exhibit shortcomings in aligning with human learning, robustness, and scalability. Addressing this, we introduce ClusterComm, a fully decentralized MARL framework where agents communicate discretely without a central control unit. ClusterComm utilizes Mini-Batch-K-Means clustering on the last hidden layer's activations of an agent's policy network, translating them into discrete messages. This approach outperforms no communication and competes favorably with unbounded, continuous communication and hence poses a simple yet effective strategy for enhancing collaborative task-solving in MARL.

cs.AI↗

arXiv2023

Attention-Based Recurrence for Multi-Agent Reinforcement Learning under Stochastic Partial Observability

Stochastic partial observability poses a major challenge for decentralized coordination in multi-agent reinforcement learning but is largely neglected in state-of-the-art research due to a strong focus on state-based centralized training for decentralized execution (CTDE) and benchmarks that lack sufficient stochasticity like StarCraft Multi-Agent Challenge (SMAC). In this paper, we propose Attention-based Embeddings of Recurrence In multi-Agent Learning (AERIAL) to approximate value functions under stochastic partial observability. AERIAL replaces the true state with a learned representation of multi-agent recurrence, considering more accurate information about decentralized agent decisions than state-based CTDE. We then introduce MessySMAC, a modified version of SMAC with stochastic observations and higher variance in initial states, to provide a more general and configurable benchmark regarding stochastic partial observability. We evaluate AERIAL in Dec-Tiger as well as in a variety of SMAC and MessySMAC maps, and compare the results with state-based CTDE. Furthermore, we evaluate the robustness of AERIAL and state-based CTDE against various stochasticity configurations in MessySMAC.

cs.MA↗

arXiv2023

Towards Efficient Quantum Anomaly Detection: One-Class SVMs using Variable Subsampling and Randomized Measurements

Quantum computing, with its potential to enhance various machine learning tasks, allows significant advancements in kernel calculation and model precision. Utilizing the one-class Support Vector Machine alongside a quantum kernel, known for its classically challenging representational capacity, notable improvements in average precision compared to classical counterparts were observed in previous studies. Conventional calculations of these kernels, however, present a quadratic time complexity concerning data size, posing challenges in practical applications. To mitigate this, we explore two distinct approaches: utilizing randomized measurements to evaluate the quantum kernel and implementing the variable subsampling ensemble method, both targeting linear time complexity. Experimental results demonstrate a substantial reduction in training and inference times by up to 95\% and 25\% respectively, employing these methods. Although unstable, the average precision of randomized measurements discernibly surpasses that of the classical Radial Basis Function kernel, suggesting a promising direction for further research in scalable, efficient quantum computing applications in machine learning.

quant-ph↗

arXiv2023

Improving Parameter Training for VQEs by Sequential Hamiltonian Assembly

A central challenge in quantum machine learning is the design and training of parameterized quantum circuits (PQCs). Similar to deep learning, vanishing gradients pose immense problems in the trainability of PQCs, which have been shown to arise from a multitude of sources. One such cause are non-local loss functions, that demand the measurement of a large subset of involved qubits. To facilitate the parameter training for quantum applications using global loss functions, we propose a Sequential Hamiltonian Assembly, which iteratively approximates the loss function using local components. Aiming for a prove of principle, we evaluate our approach using Graph Coloring problem with a Varational Quantum Eigensolver (VQE). Simulation results show, that our approach outperforms conventional parameter training by 29.99% and the empirical state of the art, Layerwise Learning, by 5.12% in the mean accuracy. This paves the way towards locality-aware learning techniques, allowing to evade vanishing gradients for a large class of practically relevant problems.

quant-ph↗

arXiv2023

Benchmarking Quantum Surrogate Models on Scarce and Noisy Data

Surrogate models are ubiquitously used in industry and academia to efficiently approximate given black box functions. As state-of-the-art methods from classical machine learning frequently struggle to solve this problem accurately for the often scarce and noisy data sets in practical applications, investigating novel approaches is of great interest. Motivated by recent theoretical results indicating that quantum neural networks (QNNs) have the potential to outperform their classical analogs in the presence of scarce and noisy data, we benchmark their qualitative performance for this scenario empirically. Our contribution displays the first application-centered approach of using QNNs as surrogate models on higher dimensional, real world data. When compared to a classical artificial neural network with a similar number of parameters, our QNN demonstrates significantly better results for noisy and scarce data, and thus motivates future work to explore this potential quantum advantage in surrogate modelling. Finally, we demonstrate the performance of current NISQ hardware experimentally and estimate the gate fidelities necessary to replicate our simulation results.

quant-ph↗

arXiv2023

Efficient and Accurate Tree Detection from 3D Point Clouds through Paid Crowdsourcing

Accurate tree detection is of growing importance in applications such as urban planning, forest inventory, and environmental monitoring. In this article, we present an approach to creating tree maps by annotating them in 3D point clouds. Point cloud representations allow the precise identification of tree positions, particularly stem locations, and their heights. Our method leverages human computational power through paid crowdsourcing, employing a web tool designed to enable even non-experts to effectively tackle the task. The primary focus of this paper is to discuss the web tool's development and strategies to ensure high-quality tree annotations despite encountering noise in the crowdsourced data. Following our methodology, we achieve quality measures surpassing 90% for various challenging test sets of diverse complexities. We emphasize that our tree map creation process, including initial point cloud collection, can be completed within 1-2 days.

cs.IR↗

arXiv2023

Improving Primate Sounds Classification using Binary Presorting for Deep Learning

In the field of wildlife observation and conservation, approaches involving machine learning on audio recordings are becoming increasingly popular. Unfortunately, available datasets from this field of research are often not optimal learning material; Samples can be weakly labeled, of different lengths or come with a poor signal-to-noise ratio. In this work, we introduce a generalized approach that first relabels subsegments of MEL spectrogram representations, to achieve higher performances on the actual multi-class classification tasks. For both the binary pre-sorting and the classification, we make use of convolutional neural networks (CNN) and various data-augmentation techniques. We showcase the results of this approach on the challenging \textit{ComparE 2021} dataset, with the task of classifying between different primate species sounds, and report significantly higher Accuracy and UAR scores in contrast to comparatively equipped model baselines.

cs.SD↗

arXiv2023

Weight Re-Mapping for Variational Quantum Algorithms

Inspired by the remarkable success of artificial neural networks across a broad spectrum of AI tasks, variational quantum circuits (VQCs) have recently seen an upsurge in quantum machine learning applications. The promising outcomes shown by VQCs, such as improved generalization and reduced parameter training requirements, are attributed to the robust algorithmic capabilities of quantum computing. However, the current gradient-based training approaches for VQCs do not adequately accommodate the fact that trainable parameters (or weights) are typically used as angles in rotational gates. To address this, we extend the concept of weight re-mapping for VQCs, as introduced by Kölle et al. (2023). This approach unambiguously maps the weights to an interval of length $2π$, mirroring data rescaling techniques in conventional machine learning that have proven to be highly beneficial in numerous scenarios. In our study, we employ seven distinct weight re-mapping functions to assess their impact on eight classification datasets, using variational classifiers as a representative example. Our results indicate that weight re-mapping can enhance the convergence speed of the VQC. We assess the efficacy of various re-mapping functions across all datasets and measure their influence on the VQC's average performance. Our findings indicate that weight re-mapping not only consistently accelerates the convergence of VQCs, regardless of the specific re-mapping function employed, but also significantly increases accuracy in certain cases.

quant-ph↗

arXiv2023

Evidence that PUBO outperforms QUBO when solving continuous optimization problems with the QAOA

Quantum computing provides powerful algorithmic tools that have been shown to outperform established classical solvers in specific optimization tasks. A core step in solving optimization problems with known quantum algorithms such as the Quantum Approximate Optimization Algorithm (QAOA) is the problem formulation. While quantum optimization has historically centered around Quadratic Unconstrained Optimization (QUBO) problems, recent studies show, that many combinatorial problems such as the TSP can be solved more efficiently in their native Polynomial Unconstrained Optimization (PUBO) forms. As many optimization problems in practice also contain continuous variables, our contribution investigates the performance of the QAOA in solving continuous optimization problems when using PUBO and QUBO formulations. Our extensive evaluation on suitable benchmark functions, shows that PUBO formulations generally yield better results, while requiring less qubits. As the multi-qubit interactions needed for the PUBO variant have to be decomposed using the hardware gates available, i.e., currently single- and two-qubit gates, the circuit depth of the PUBO approach outscales its QUBO alternative roughly linearly in the order of the objective function. However, incorporating the planned addition of native multi-qubit gates such as the global Molmer-Sorenson gate, our experiments indicate that PUBO outperforms QUBO for higher order continuous optimization problems in general.

quant-ph↗

arXiv2023

Constructing Organism Networks from Collaborative Self-Replicators

We introduce organism networks, which function like a single neural network but are composed of several neural particle networks; while each particle network fulfils the role of a single weight application within the organism network, it is also trained to self-replicate its own weights. As organism networks feature vastly more parameters than simpler architectures, we perform our initial experiments on an arithmetic task as well as on simplified MNIST-dataset classification as a collective. We observe that individual particle networks tend to specialise in either of the tasks and that the ones fully specialised in the secondary task may be dropped from the network without hindering the computational accuracy of the primary task. This leads to the discovery of a novel pruning-strategy for sparse neural networks

cs.NE↗

arXiv2023

Improving Convergence for Quantum Variational Classifiers using Weight Re-Mapping

In recent years, quantum machine learning has seen a substantial increase in the use of variational quantum circuits (VQCs). VQCs are inspired by artificial neural networks, which achieve extraordinary performance in a wide range of AI tasks as massively parameterized function approximators. VQCs have already demonstrated promising results, for example, in generalization and the requirement for fewer parameters to train, by utilizing the more robust algorithmic toolbox available in quantum computing. A VQCs' trainable parameters or weights are usually used as angles in rotational gates and current gradient-based training methods do not account for that. We introduce weight re-mapping for VQCs, to unambiguously map the weights to an interval of length $2π$, drawing inspiration from traditional ML, where data rescaling, or normalization techniques have demonstrated tremendous benefits in many circumstances. We employ a set of five functions and evaluate them on the Iris and Wine datasets using variational classifiers as an example. Our experiments show that weight re-mapping can improve convergence in all tested settings. Additionally, we were able to demonstrate that weight re-mapping increased test accuracy for the Wine dataset by $10\%$ over using unmodified weights.

quant-ph↗

arXiv2023

Learning to Participate through Trading of Reward Shares

Enabling autonomous agents to act cooperatively is an important step to integrate artificial intelligence in our daily lives. While some methods seek to stimulate cooperation by letting agents give rewards to others, in this paper we propose a method inspired by the stock market, where agents have the opportunity to participate in other agents' returns by acquiring reward shares. Intuitively, an agent may learn to act according to the common interest when being directly affected by the other agents' rewards. The empirical results of the tested general-sum Markov games show that this mechanism promotes cooperative policies among independently trained agents in social dilemma situations. Moreover, as demonstrated in a temporally and spatially extended domain, participation can lead to the development of roles and the division of subtasks between the agents.

cs.MA↗

arXiv2023

Compression of GPS Trajectories using Autoencoders

The ubiquitous availability of mobile devices capable of location tracking led to a significant rise in the collection of GPS data. Several compression methods have been developed in order to reduce the amount of storage needed while keeping the important information. In this paper, we present an lstm-autoencoder based approach in order to compress and reconstruct GPS trajectories, which is evaluated on both a gaming and real-world dataset. We consider various compression ratios and trajectory lengths. The performance is compared to other trajectory compression algorithms, i.e., Douglas-Peucker. Overall, the results indicate that our approach outperforms Douglas-Peucker significantly in terms of the discrete Fréchet distance and dynamic time warping. Furthermore, by reconstructing every point lossy, the proposed methodology offers multiple advantages over traditional methods.

cs.LG↗

arXiv2022

VoronoiPatches: Evaluating A New Data Augmentation Method

Overfitting is a problem in Convolutional Neural Networks (CNN) that causes poor generalization of models on unseen data. To remediate this problem, many new and diverse data augmentation methods (DA) have been proposed to supplement or generate more training data, and thereby increase its quality. In this work, we propose a new data augmentation algorithm: VoronoiPatches (VP). We primarily utilize non-linear recombination of information within an image, fragmenting and occluding small information patches. Unlike other DA methods, VP uses small convex polygon-shaped patches in a random layout to transport information around within an image. Sudden transitions created between patches and the original image can, optionally, be smoothed. In our experiments, VP outperformed current DA methods regarding model variance and overfitting tendencies. We demonstrate data augmentation utilizing non-linear re-combination of information within images, and non-orthogonal shapes and structures improves CNN model robustness on unseen data.

cs.CV↗

arXiv2022

Decentralized scheduling through an adaptive, trading-based multi-agent system

In multi-agent reinforcement learning systems, the actions of one agent can have a negative impact on the rewards of other agents. One way to combat this problem is to let agents trade their rewards amongst each other. Motivated by this, this work applies a trading approach to a simulated scheduling environment, where the agents are responsible for the assignment of incoming jobs to compute cores. In this environment, reinforcement learning agents learn to trade successfully. The agents can trade the usage right of computational cores to process high-priority, high-reward jobs faster than low-priority, low-reward jobs. However, due to combinatorial effects, the action and observation spaces of a simple reinforcement learning agent in this environment scale exponentially with key parameters of the problem size. However, the exponential scaling behavior can be transformed into a linear one if the agent is split into several independent sub-units. We further improve this distributed architecture using agent-internal parameter sharing. Moreover, it can be extended to set the exchange prices autonomously. We show that in our scheduling environment, the advantages of a distributed agent architecture clearly outweigh more aggregated approaches. We demonstrate that the distributed agent architecture becomes even more performant using agent-internal parameter sharing. Finally, we investigate how two different reward functions affect autonomous pricing and the corresponding scheduling.

cs.AI↗

arXiv2021

Quantifying Multimodality in World Models

Model-based Deep Reinforcement Learning (RL) assumes the availability of a model of an environment's underlying transition dynamics. This model can be used to predict future effects of an agent's possible actions. When no such model is available, it is possible to learn an approximation of the real environment, e.g. by using generative neural networks, sometimes also called World Models. As most real-world environments are stochastic in nature and the transition dynamics are oftentimes multimodal, it is important to use a modelling technique that is able to reflect this multimodal uncertainty. In order to safely deploy such learning systems in the real world, especially in an industrial context, it is paramount to consider these uncertainties. In this work, we analyze existing and propose new metrics for the detection and quantification of multimodal uncertainty in RL based World Models. The correct modelling & detection of uncertain future states lays the foundation for handling critical situations in a safe way, which is a prerequisite for deploying RL systems in real-world settings.

cs.LG↗