arXiv subjects

Tobias Jacobs

Publications and source records attributed to Tobias Jacobs.

11 recordsLinked to original sources

arXiv2024

Interactive Ontology Matching with Cost-Efficient Learning

The creation of high-quality ontologies is crucial for data integration and knowledge-based reasoning, specifically in the context of the rising data economy. However, automatic ontology matchers are often bound to the heuristics they are based on, leaving many matches unidentified. Interactive ontology matching systems involving human experts have been introduced, but they do not solve the fundamental issue of flexibly finding additional matches outside the scope of the implemented heuristics, even though this is highly demanded in industrial settings. Active machine learning methods appear to be a promising path towards a flexible interactive ontology matcher. However, off-the-shelf active learning mechanisms suffer from low query efficiency due to extreme class imbalance, resulting in a last-mile problem where high human effort is required to identify the remaining matches. To address the last-mile problem, this work introduces DualLoop, an active learning method tailored to ontology matching. DualLoop offers three main contributions: (1) an ensemble of tunable heuristic matchers, (2) a short-term learner with a novel query strategy adapted to highly imbalanced data, and (3) long-term learners to explore potential matches by creating and tuning new heuristics. We evaluated DualLoop on three datasets of varying sizes and domains. Compared to existing active learning methods, we consistently achieved better F1 scores and recall, reducing the expected query cost spent on finding 90% of all matches by over 50%. Compared to traditional interactive ontology matchers, we are able to find additional, last-mile matches. Finally, we detail the successful deployment of our approach within an actual product and report its operational performance results within the Architecture, Engineering, and Construction (AEC) industry sector, showcasing its practical value and efficiency.

cs.DB↗

arXiv2022

ProcK: Machine Learning for Knowledge-Intensive Processes

We present a novel methodology to build powerful predictive process models. Our method, denoted ProcK (Process & Knowledge), relies not only on sequential input data in the form of event logs, but can learn to use a knowledge graph to incorporate information about the attribute values of the events and their mutual relationships. The idea is realized by mapping event attributes to nodes of a knowledge graph and training a sequence model alongside a graph neural network in an end-to-end fashion. This hybrid approach substantially enhances the flexibility and applicability of predictive process monitoring, as both the static and dynamic information residing in the databases of organizations can be directly taken as input data. We demonstrate the potential of ProcK by applying it to a number of predictive process monitoring tasks, including tasks with knowledge graphs available as well as an existing process monitoring benchmark where no such graph is given. The experiments provide evidence that our methodology achieves state-of-the-art performance and improves predictive power when a knowledge graph is available.

cs.LG↗

arXiv2021

ITS and Real Time Cross Border Logistic Operations Optimization

Moving parcels from origin to destination should not require a lot of re-planning. However, the vast number of shipments and destinations, which need to be re-aligned in real-time due to various external factors makes the delivery process a complex issue to tackle. Anticipating the impact of external factors though can provide more robust logistic plans which are resilient to changes. The work described in this paper, was carried out in the EU-funded COG-LO project and addresses the issue of parcel delivery across the road network making use of context-awareness information as an input for the optimization operations. A positive impact derived from the implementation of these services is expected due to complex event detection, context awareness and decision support at both local and global level of logistics operations.

eess.SY↗

arXiv2020

The Combinatorial Multi-Bandit Problem and its Application to Energy Management

We study a Combinatorial Multi-Bandit Problem motivated by applications in energy systems management. Given multiple probabilistic multi-arm bandits with unknown outcome distributions, the task is to optimize the value of a combinatorial objective function mapping the vector of individual bandit outcomes to a single scalar reward. Unlike in single-bandit problems with multi-dimensional action space, the outcomes of the individual bandits are observable in our setting and the objective function is known. Guided by the hypothesis that individual observability enables better trade-offs between exploration and exploitation, we generalize the lower regret bound for single bandits, showing that indeed for multiple bandits it admits parallelized exploration. For our energy management application we propose a range of algorithms that combine exploration principles for multi-arm bandits with mathematical programming. In an experimental study we demonstrate the effectiveness of our approach to learn action assignments for 150 bandits, each having 24 actions, within a horizon of 365 episodes.

cs.LG↗

arXiv2020

Large-Scale Cargo Distribution

This study focuses on the design and development of methods for generating cargo distribution plans for large-scale logistics networks. It uses data from three large logistics operators while focusing on cross border logistics operations using one large graph. The approach uses a three-step methodology to first represent the logistic infrastructure as a graph, then partition the graph into smaller size regions, and finally generate cargo distribution plans for each individual region. The initial graph representation has been extracted from regional graphs by spectral clustering and is then further used for computing the distribution plan. The approach introduces methods for each of the modelling steps. The proposed approach on using regionalization of large logistics infrastructure for generating partial plans, enables scaling to thousands of drop-off locations. Results also show that the proposed approach scales better than the state-of-the-art, while preserving the quality of the solution. Our methodology is suited to address the main challenge in transforming rigid large logistics infrastructure into dynamic, just-in-time, and point-to-point delivery-oriented logistics operations.

cs.AI↗

arXiv2019

A Study of the Learning Progress in Neural Architecture Search Techniques

In neural architecture search, the structure of the neural network to best model a given dataset is determined by an automated search process. Efficient Neural Architecture Search (ENAS), proposed by Pham et al. (2018), has recently received considerable attention due to its ability to find excellent architectures within a comparably short search time. In this work, which is motivated by the quest to further improve the learning speed of architecture search, we evaluate the learning progress of the controller which generates the architectures in ENAS. We measure the progress by comparing the architectures generated by it at different controller training epochs, where architectures are evaluated after having re-trained them from scratch. As a surprising result, we find that the learning curves are completely flat, i.e., there is no observable progress of the controller in terms of the performance of its generated architectures. This observation is consistent across the CIFAR-10 and CIFAR-100 datasets and two different search spaces. We conclude that the high quality of the models generated by ENAS is a result of the search space design rather than the controller training, and our results indicate that one-shot architecture design is an efficient alternative to architecture search by ENAS.

cs.LG↗

arXiv2019

On the Performance of Differential Evolution for Hyperparameter Tuning

Automated hyperparameter tuning aspires to facilitate the application of machine learning for non-experts. In the literature, different optimization approaches are applied for that purpose. This paper investigates the performance of Differential Evolution for tuning hyperparameters of supervised learning algorithms for classification tasks. This empirical study involves a range of different machine learning algorithms and datasets with various characteristics to compare the performance of Differential Evolution with Sequential Model-based Algorithm Configuration (SMAC), a reference Bayesian Optimization approach. The results indicate that Differential Evolution outperforms SMAC for most datasets when tuning a given machine learning algorithm - particularly when breaking ties in a first-to-report fashion. Only for the tightest of computational budgets SMAC performs better. On small datasets, Differential Evolution outperforms SMAC by 19% (37% after tie-breaking). In a second experiment across a range of representative datasets taken from the literature, Differential Evolution scores 15% (23% after tie-breaking) more wins than SMAC.

cs.LG↗

arXiv2017

Re-designing Dynamic Content Delivery in the Light of a Virtualized Infrastructure

We explore the opportunities and design options enabled by novel SDN and NFV technologies, by re-designing a dynamic Content Delivery Network (CDN) service. Our system, named MOSTO, provides performance levels comparable to that of a regular CDN, but does not require the deployment of a large distributed infrastructure. In the process of designing the system, we identify relevant functions that could be integrated in the future Internet infrastructure. Such functions greatly simplify the design and effectiveness of services such as MOSTO. We demonstrate our system using a mixture of simulation, emulation, testbed experiments and by realizing a proof-of-concept deployment in a planet-wide commercial cloud system.

cs.NI↗

arXiv2014

A New Perspective on the Windows Scheduling Problem

The Windows Scheduling Problem, also known as the Pinwheel Problem, is to schedule periodic jobs subject to their processing frequency demands. Instances are given as a set of jobs that have to be processed infinitely often such that the time interval between two consecutive executions of the same job j is no longer than the job's given period $p_j$. The key contribution of this work is a new interpretation of the problem variant with exact periods, where the time interval between consecutive executions must be strictly $p_j$. We show that this version is equivalent to a natural combinatorial problem we call Partial Coding. Reductions in both directions can be realized in polynomial time, so that both hardness proofs and algorithms for Partial Coding transfer to Windows Scheduling. Applying this new perspective, we obtain a number of new results regarding the computational complexity of various Windows Scheduling Problem variants. We prove that even the case of one processor and unit-length jobs does not admit a pseudo-polynomial time algorithm unless SAT can be solved by a randomized method in expected quasi-polynomial time. This result also extends to the case of inexact periods, which answers a question that has remained open for more than two decades. Furthermore, we report an error found in a hardness proof previously given for the multi-machine case without machine migration, and we show that this variant reduces to the single-machine case. Finally, we prove that even with unit-length jobs the problem is co-NP-hard when jobs are allowed to migrate between machines.

cs.CC↗

arXiv2009

Simpler Proofs by Symbolic Perturbation

In analyses of algorithms, a substantial amount of effort has often to be spent on the discussion of special cases. For example, when the analysis considers the cases X Y separately, one might have to be especially careful about what happens when X=Y. On the other hand, experience tells us that when a yet unregarded special case of this kind is discovered, one nearly always finds a way to handle it. This is typically done by modifying the analysis and/or the algorithm very slightly. In this article we substantiate this observation theoretically. We concentrate on deterministic algorithms for weighted combinatorial optimization problems. A problem instance of this kind is defined by its structure and a vector of weights. The concept of a null case is introduced as set of problem instances whose weight vectors constitute a nowhere open set (or null set) in the space of all possible weight configurations. An algorithm is called robust if any null case can be disregarded in the analysis of both its solution quality and resource requirements. We show that achieving robustness is only a matter of breaking ties the right way. More specifically, we show that the concept of symbolic perturbation known from the area of geometric algorithms guarantees that no surprises will happen in null cases. We argue that for a huge class of combinatorial optimization algorithms it is easy to verify that they implicitly use symbolic perturbation for breaking ties and thus can be analyzed under the assumption that some arbitrary null case never occurs. Finally, we prove that there exists a symbolic perturbation tie breaking policy for any algorithm.

cs.DS↗

arXiv2009

On the Complexity of Searching in Trees: Average-case Minimization

We focus on the average-case analysis: A function w : V -> Z+ is given which defines the likelihood for a node to be the one marked, and we want the strategy that minimizes the expected number of queries. Prior to this paper, very little was known about this natural question and the complexity of the problem had remained so far an open question. We close this question and prove that the above tree search problem is NP-complete even for the class of trees with diameter at most 4. This results in a complete characterization of the complexity of the problem with respect to the diameter size. In fact, for diameter not larger than 3 the problem can be shown to be polynomially solvable using a dynamic programming approach. In addition we prove that the problem is NP-complete even for the class of trees of maximum degree at most 16. To the best of our knowledge, the only known result in this direction is that the tree search problem is solvable in O(|V| log|V|) time for trees with degree at most 2 (paths). We match the above complexity results with a tight algorithmic analysis. We first show that a natural greedy algorithm attains a 2-approximation. Furthermore, for the bounded degree instances, we show that any optimal strategy (i.e., one that minimizes the expected number of queries) performs at most O(Δ(T) (log |V| + log w(T))) queries in the worst case, where w(T) is the sum of the likelihoods of the nodes of T and Δ(T) is the maximum degree of T. We combine this result with a non-trivial exponential time algorithm to provide an FPTAS for trees with bounded degree.

cs.DS↗