arXiv subjects

Claudia Linnhoff-Popien

Publications and source records attributed to Claudia Linnhoff-Popien.

141 records · Page 8Linked to original sources

arXiv2020

The Holy Grail of Quantum Artificial Intelligence: Major Challenges in Accelerating the Machine Learning Pipeline

We discuss the synergetic connection between quantum computing and artificial intelligence. After surveying current approaches to quantum artificial intelligence and relating them to a formal model for machine learning processes, we deduce four major challenges for the future of quantum artificial intelligence: (i) Replace iterative training with faster quantum algorithms, (ii) distill the experience of larger amounts of data into the training process, (iii) allow quantum and classical components to be easily combined and exchanged, and (iv) build tools to thoroughly analyze whether observed benefits really stem from quantum properties of the algorithm.

quant-ph↗

arXiv2020

Insights on Training Neural Networks for QUBO Tasks

Current hardware limitations restrict the potential when solving quadratic unconstrained binary optimization (QUBO) problems via the quantum approximate optimization algorithm (QAOA) or quantum annealing (QA). Thus, we consider training neural networks in this context. We first discuss QUBO problems that originate from translated instances of the traveling salesman problem (TSP): Analyzing this representation via autoencoders shows that there is way more information included than necessary to solve the original TSP. Then we show that neural networks can be used to solve TSP instances from both QUBO input and autoencoders' hiddenstate representation. We finally generalize the approach and successfully train neural networks to solve arbitrary QUBO problems, sketching means to use neuromorphic hardware as a simulator or an additional co-processor for quantum computing.

quant-ph↗

arXiv2020

Approximate Approximation on a Quantum Annealer

Many problems of industrial interest are NP-complete, and quickly exhaust resources of computational devices with increasing input sizes. Quantum annealers (QA) are physical devices that aim at this class of problems by exploiting quantum mechanical properties of nature. However, they compete with efficient heuristics and probabilistic or randomised algorithms on classical machines that allow for finding approximate solutions to large NP-complete problems. While first implementations of QA have become commercially available, their practical benefits are far from fully explored. To the best of our knowledge, approximation techniques have not yet received substantial attention. In this paper, we explore how problems' approximate versions of varying degree can be systematically constructed for quantum annealer programs, and how this influences result quality or the handling of larger problem instances on given set of qubits. We illustrate various approximation techniques on both, simulations and real QA hardware, on different seminal problems, and interpret the results to contribute towards a better understanding of the real-world power and limitations of current-state and future quantum computing.

quant-ph↗

arXiv2020

Soccer Team Vectors

In this work we present STEVE - Soccer TEam VEctors, a principled approach for learning real valued vectors for soccer teams where similar teams are close to each other in the resulting vector space. STEVE only relies on freely available information about the matches teams played in the past. These vectors can serve as input to various machine learning tasks. Evaluating on the task of team market value estimation, STEVE outperforms all its competitors. Moreover, we use STEVE for similarity search and to rank soccer teams.

cs.LG↗

arXiv2020

Optimizing Geometry Compression using Quantum Annealing

The compression of geometry data is an important aspect of bandwidth-efficient data transfer for distributed 3d computer vision applications. We propose a quantum-enabled lossy 3d point cloud compression pipeline based on the constructive solid geometry (CSG) model representation. Key parts of the pipeline are mapped to NP-complete problems for which an efficient Ising formulation suitable for the execution on a Quantum Annealer exists. We describe existing Ising formulations for the maximum clique search problem and the smallest exact cover problem, both of which are important building blocks of the proposed compression pipeline. Additionally, we discuss the properties of the overall pipeline regarding result optimality and described Ising formulations.

quant-ph↗

arXiv2019

Uncertainty-Based Out-of-Distribution Classification in Deep Reinforcement Learning

Robustness to out-of-distribution (OOD) data is an important goal in building reliable machine learning systems. Especially in autonomous systems, wrong predictions for OOD inputs can cause safety critical situations. As a first step towards a solution, we consider the problem of detecting such data in a value-based deep reinforcement learning (RL) setting. Modelling this problem as a one-class classification problem, we propose a framework for uncertainty-based OOD classification: UBOOD. It is based on the effect that an agent's epistemic uncertainty is reduced for situations encountered during training (in-distribution), and thus lower than for unencountered (OOD) situations. Being agnostic towards the approach used for estimating epistemic uncertainty, combinations with different uncertainty estimation methods, e.g. approximate Bayesian inference methods or ensembling techniques are possible. We further present a first viable solution for calculating a dynamic classification threshold, based on the uncertainty distribution of the training data. Evaluation shows that the framework produces reliable classification results when combined with ensemble-based estimators, while the combination with concrete dropout-based estimators fails to reliably detect OOD situations. In summary, UBOOD presents a viable approach for OOD classification in deep RL settings by leveraging the epistemic uncertainty of the agent's value function.

cs.LG↗

arXiv2019

Difficulty Classification of Mountainbike Downhill Trails utilizing Deep Neural Networks

The difficulty of mountainbike downhill trails is a subjective perception. However, sports-associations and mountainbike park operators attempt to group trails into different levels of difficulty with scales like the Singletrail-Skala (S0-S5) or colored scales (blue, red, black, ...) as proposed by The International Mountain Bicycling Association. Inconsistencies in difficulty grading occur due to the various scales, different people grading the trails, differences in topography, and more. We propose an end-to-end deep learning approach to classify trails into three difficulties easy, medium, and hard by using sensor data. With mbientlab Meta Motion r0.2 sensor units, we record accelerometer- and gyroscope data of one rider on multiple trail segments. A 2D convolutional neural network is trained with a stacked and concatenated representation of the aforementioned data as its input. We run experiments with five different sample- and five different kernel sizes and achieve a maximum Sparse Categorical Accuracy of 0.9097. To the best of our knowledge, this is the first work targeting computational difficulty classification of mountainbike downhill trails.

cs.CV↗

arXiv2019

A Hybrid Solution Method for the Capacitated Vehicle Routing Problem Using a Quantum Annealer

The Capacitated Vehicle Routing Problem (CVRP) is an NP-optimization problem (NPO) that has been of great interest for decades for both, science and industry. The CVRP is a variant of the vehicle routing problem characterized by capacity constrained vehicles. The aim is to plan tours for vehicles to supply a given number of customers as efficiently as possible. The problem is the combinatorial explosion of possible solutions, which increases superexponentially with the number of customers. Classical solutions provide good approximations to the globally optimal solution. D-Wave's quantum annealer is a machine designed to solve optimization problems. This machine uses quantum effects to speed up computation time compared to classic computers. The problem on solving the CVRP on the quantum annealer is the particular formulation of the optimization problem. For this, it has to be mapped onto a quadratic unconstrained binary optimization (QUBO) problem. Complex optimization problems such as the CVRP can be translated to smaller subproblems and thus enable a sequential solution of the partitioned problem. This work presents a quantum-classic hybrid solution method for the CVRP. It clarifies whether the implemenation of such a method pays off in comparison to existing classical solution methods regarding computation time and solution quality. Several approaches to solving the CVRP are elaborated, the arising problems are discussed, and the results are evaluated in terms of solution quality and computation time.

quant-ph↗

arXiv2019

Memory Bounded Open-Loop Planning in Large POMDPs using Thompson Sampling

State-of-the-art approaches to partially observable planning like POMCP are based on stochastic tree search. While these approaches are computationally efficient, they may still construct search trees of considerable size, which could limit the performance due to restricted memory resources. In this paper, we propose Partially Observable Stacked Thompson Sampling (POSTS), a memory bounded approach to open-loop planning in large POMDPs, which optimizes a fixed size stack of Thompson Sampling bandits. We empirically evaluate POSTS in four large benchmark problems and compare its performance with different tree-based approaches. We show that POSTS achieves competitive performance compared to tree-based open-loop planning and offers a performance-memory tradeoff, making it suitable for partially observable planning with highly restricted computational and memory resources.

cs.AI↗

arXiv2019

Emergent Escape-based Flocking Behavior using Multi-Agent Reinforcement Learning

In nature, flocking or swarm behavior is observed in many species as it has beneficial properties like reducing the probability of being caught by a predator. In this paper, we propose SELFish (Swarm Emergent Learning Fish), an approach with multiple autonomous agents which can freely move in a continuous space with the objective to avoid being caught by a present predator. The predator has the property that it might get distracted by multiple possible preys in its vicinity. We show that this property in interaction with self-interested agents which are trained with reinforcement learning to solely survive as long as possible leads to flocking behavior similar to Boids, a common simulation for flocking behavior. Furthermore we present interesting insights in the swarming behavior and in the process of agents being caught in our modeled environment.

cs.MA↗

arXiv2019

Assessing Solution Quality of 3SAT on a Quantum Annealing Platform

When solving propositional logic satisfiability (specifically 3SAT) using quantum annealing, we analyze the effect the difficulty of different instances of the problem has on the quality of the answer returned by the quantum annealer. A high-quality response from the annealer in this case is defined by a high percentage of correct solutions among the returned answers. We show that the phase transition regarding the computational complexity of the problem, which is well-known to occur for 3SAT on classical machines (where it causes a detrimental increase in runtime), persists in some form (but possibly to a lesser extent) for quantum annealing.

cs.ET↗

arXiv2019

Distributed Policy Iteration for Scalable Approximation of Cooperative Multi-Agent Policies

Decision making in multi-agent systems (MAS) is a great challenge due to enormous state and joint action spaces as well as uncertainty, making centralized control generally infeasible. Decentralized control offers better scalability and robustness but requires mechanisms to coordinate on joint tasks and to avoid conflicts. Common approaches to learn decentralized policies for cooperative MAS suffer from non-stationarity and lacking credit assignment, which can lead to unstable and uncoordinated behavior in complex environments. In this paper, we propose Strong Emergent Policy approximation (STEP), a scalable approach to learn strong decentralized policies for cooperative MAS with a distributed variant of policy iteration. For that, we use function approximation to learn from action recommendations of a decentralized multi-agent planning algorithm. STEP combines decentralized multi-agent planning with centralized learning, only requiring a generative model for distributed black box optimization. We experimentally evaluate STEP in two challenging and stochastic domains with large state and joint action spaces and show that STEP is able to learn stronger policies than standard multi-agent reinforcement learning algorithms, when combining multi-agent open-loop planning with centralized function approximation. The learned policies can be reintegrated into the multi-agent planning process to further improve performance.

cs.AI↗

arXiv2019

Uncertainty-Based Out-of-Distribution Detection in Deep Reinforcement Learning

We consider the problem of detecting out-of-distribution (OOD) samples in deep reinforcement learning. In a value based reinforcement learning setting, we propose to use uncertainty estimation techniques directly on the agent's value estimating neural network to detect OOD samples. The focus of our work lies in analyzing the suitability of approximate Bayesian inference methods and related ensembling techniques that generate uncertainty estimates. Although prior work has shown that dropout-based variational inference techniques and bootstrap-based approaches can be used to model epistemic uncertainty, the suitability for detecting OOD samples in deep reinforcement learning remains an open question. Our results show that uncertainty estimation can be used to differentiate in- from out-of-distribution samples. Over the complete training process of the reinforcement learning agents, bootstrap-based approaches tend to produce more reliable epistemic uncertainty estimates, when compared to dropout-based approaches.

cs.LG↗

arXiv2018

Inheritance-Based Diversity Measures for Explicit Convergence Control in Evolutionary Algorithms

Diversity is an important factor in evolutionary algorithms to prevent premature convergence towards a single local optimum. In order to maintain diversity throughout the process of evolution, various means exist in literature. We analyze approaches to diversity that (a) have an explicit and quantifiable influence on fitness at the individual level and (b) require no (or very little) additional domain knowledge such as domain-specific distance functions. We also introduce the concept of genealogical diversity in a broader study. We show that employing these approaches can help evolutionary algorithms for global optimization in many cases.

cs.NE↗

arXiv2017

SensX: About Sensing and Assessment of Complex Human Motion

The great success of wearables and smartphone apps for provision of extensive physical workout instructions boosts a whole industry dealing with consumer oriented sensors and sports equipment. But with these opportunities there are also new challenges emerging. The unregulated distribution of instructions about ambitious exercises enables unexperienced users to undertake demanding workouts without professional supervision which may lead to suboptimal training success or even serious injuries. We believe, that automated supervision and realtime feedback during a workout may help to solve these issues. Therefore we introduce four fundamental steps for complex human motion assessment and present SensX, a sensor-based architecture for monitoring, recording, and analyzing complex and multi-dimensional motion chains. We provide the results of our preliminary study encompassing 8 different body weight exercises, 20 participants, and more than 9,220 recorded exercise repetitions. Furthermore, insights into SensXs classification capabilities and the impact of specific sensor configurations onto the analysis process are given.

cs.CY↗