Monitoring hardware utilization when training on GPUs in distributed machine learning

Large­scale machine learning frameworks can accelerate training of a neural network by per­ forming distributed training on a cluster using multiple GPUs per node and multiple nodes. Because distributed training on a cluster involves many nodes which need to communicate and load and exchange data, a machine learning framework may at certain...

Privacy-preserving Cross-Domain Recommender System for Healthcare Applications

Healthcare recommender systems emerged to help patients make better decisions for their health, leveraging the vast amount of data and patient experience. One type of this system focuses on recommending the most appropriate physician based on previous patient feedback in the form of ratings. Such advice can be challenging to generate for new...

A Heuristic Algorithm for the Flexible Job Shop Problem with Changeover Times

In this paper a heuristic algorithm is described that can constructively produce solutions to a variant of the Flexible Job Shop Problem (FJSP) that introduces changeover times between each pair of two operations consecutively performed on a machine. The performance of the heuristic algorithm is compared to the performance of an exact solver....

Ant Colony Optimization for DSM's Flexible Job Shop Problem

This papers examines an ant colony optimization approach for solving a specific variant of the Flexible Job Shop Problem faced by the Dutch chemistry company DSM. Jobs consisting of operations on a specific enzyme need to be scheduled as efficiently as possible on groups of available machines. The most interesting requirement upon the general...

Flexible Job Shop Optimization with Simulated Annealing

In this paper, a Simulated Annealing (SA) implementation for a Flexible Job Shop Problem (FJSP), with change-over time, is presented. <br/> This implementation is compared to a Mixed Integer Linear Programming (MILP) optimization, to compare performances. <br/> The SA algorithm starts with creating the first schedule with Global Selection. <br/>...

Applying hierarchical tabu search to an adapted version of the flexible job shop problem

The scheduling departments of batch manufacturing plants have to repeatedly solve a complex scheduling problem for the operation of their production lines. This problem can be modeled as a flexible job shop problem (FJSP) in which a set of operations has to be assigned to a set of machines and then the order of operations on each machine has to...

Discovering the misconceptions that influence learning of Machine Learning

The purpose of the study is to investigate and evaluate the most common misunderstandings that are associated with the subject of Machine Learning, and then to make a direct comparison between the findings of the study and the Machine Learning course that is currently being provided by the Delft University of Technology as part of their Bachelor...

A Guidline for Creating Assessments in Machine Learning Education

Even though machine learning field is growing rapidly, research on education of machine learning is scarce. In this paper a research about creating assessments in the machine learning’s context is presented. The aim of the research is to answer how to design assessments that reliably show progress on a module in machine learning. Learning...

Competences in Machine Learning: The order of competences that students need to learn in ML

Machine learning is becoming more and more applied within business and academia alike. This has led researchers to look inwards and discuss whether the current way of teaching and learning machine learning is the right way. Within this train of thought, one must investigate the characteristics of teaching and learning. Namely, what are the...

Benchmarking Checkpoint-based Fault Tolerance Algorithms in Stateful Stream Processing

Major advances in the fault tolerance of distributed stream processing systems provided the systems with the capacity to produce strictly consistent results under failures. Consistent fault tolerance has been one of the catalysts fueling the maturity of streaming systems and boosting their widespread adoption not only for analytics use cases,...

Compiling with Command Trees

Compilers translate high-level source code into low-level machine code. To represent source code a compiler uses a language called the intermediate representation (IR). An IR for the compilation of functional languages is continuation-passing style (CPS). It provides convenient abstractions for both data flow and control flow. However, CPS...