Presentation at COPA 2019

Ola Spjuth, Co-PI in the HASTE project, presented two accepted HASTE-papers at the [8th Symposium on Conformal and Probabilistic Prediction with Applications](http://clrc.rhul.ac.uk/copa2019) in Varna, Bulgaria on 9-11 Sept 2019. The two papers below are now published in [Proceedings of Machine Learning Research (PMLR) volume 105](https://proceedings.mlr.press/v105/).

Paper 1: Split Knowledge Transfer in Learning Under Privileged Information Framework

Gauraha, N., Söderdahl, F. and Spjuth, O.
Split Knowledge Transfer in Learning Under Privileged Information Framework. 
Proceedings of Machine Learning Research (PMLR). 105, 43-52. (2019).
ABSTRACT
Learning Under Privileged Information (LUPI) enables the inclusion of additional (privileged) information when training machine learning models, data that is not available when making predictions. The methodology has been successfully applied to a diverse set of problems from various fields. SVM+ was the first realization of the LUPI paradigm which showed fast convergence but did not scale well. To address the scalability issue, knowledge transfer approaches were proposed to estimate privileged information from standard features in order to construct improved decision rules. Most available knowledge transfer methods use regression techniques and the same data for approximating the privileged features as for learning the transfer function. Inspired by the cross-validation approach, we propose to partition the training data into K folds and use each fold for learning a transfer function and the remaining folds for approximations of privileged features—we refer to this as split knowledge transfer. We evaluate the method using four different experimental setups comprising one synthetic and three real datasets. The results indicate that our approach leads to improved accuracy as compared to LUPI with standard knowledge transfer.

Paper 2: Combining Prediction Intervals on Multi-Source Non-Disclosed Regression Datasets

Spjuth O., Brännström R.C., Carlsson L. and Gauraha, N.
Combining Prediction Intervals on Multi-Source Non-Disclosed Regression Datasets.
Proceedings of Machine Learning Research (PMLR). 105, 53-65. (2019).
ABSTRACT
Conformal Prediction is a framework that produces prediction intervals based on the output from a machine learning algorithm. In this paper, we explore the case when training data is made up of multiple parts available in different sources that cannot be pooled. We here consider the regression case and propose a method where a conformal predictor is trained on each data source independently, and where the prediction intervals are then combined into a single interval. We call the approach Non-Disclosed Conformal Prediction (NDCP), and we evaluate it on a regression dataset from the UCI machine learning repository using support vector regression as the underlying machine learning algorithm, with a varying number of data sources and sizes. The results show that the proposed method produces conservatively valid prediction intervals, and while we cannot retain the same efficiency as when all data is used, efficiency is improved through the proposed approach as compared to predicting using a single arbitrarily chosen source.

Presentation at DBDM/CCGrid

I’ve just presented the paper “Adapting The Secretary Hiring Problem for Optimal Hot-Cold Tier Placement under Top-K Workloads” at DBDM, CCGrid here in Larnaca, Cyprus.

 

The paper examines analytic solutions to optimization problems related to tiered/hierarchical storage under Top-K queries with HASTE, and its relation to the classic discrete optimization ‘Secretary Hiring Problem’.

Pre-Preprint

PhD students’ visit to Astra Zeneca, Gothenburg (April 15-17)

PhD students Håkan Wieslander, Phil Harrison and Ankit Gupta visited Astra Zeneca, hosted by Johan Karlsson and Alan Sabirsh. They had three intense days in the lab getting the high-throughput microscope to talk to the HASTE code. It’s not every day a computer scientist gets to dress up in a lab coat!
Johan explains the workings of the microscope
It’s not easy to debug
What’s an intense coding session without some finger-pointing

Ankit Gupta joins HASTE team as PhD student

We welcome Ankit Gupta as new PhD Student in the Wählby Lab at the Department of Information Technology, Uppsala University.

Ankit obtained his Bachelor’s in Electrical Engineering at Indian Institute of Technology Indore in 2014. Then, he completed his Masters in Medical Imaging and Informatics at Indian Institute of Technology Kharagpur in 2017. Before moving to Uppsala, he was employed as Research Engineer at the University of Bern where he worked on developing a video-based instrument tracking system in stereoscopic laparoscopic surgery.

About the PhD project within HASTE:  

Within the project, he will work on developing measurements for the early detection of informative data from large-scale spatial and temporal experiments.

Successful HASTE ‘all hands’ at Uppsala (Nov 7-9)

Johan makes a start on the fika…
Everyone presented their latest work, and discussed the latest image datasets from AstraZeneca and Vironova. During the software workshop session, we discussed linking the HASTE cloud pipeline to the Vironova MiniTEM.

Thanks to: Carolina Wählby, Ola Spjuth, Andreas Hellander, Ida-Maria Sintorn, Alan Sabirsh, Ernst Ahlberg Helgee, Johan Karlsson, Håkan Wieslander, Philip Harrison, Salman Toor, Ben Blamey, Håkan Öhrn, Markus M. Hilscher, Niharika Gauraha, Magnus Larsson, Oliver Stein, Andy Ishak

HASTE project featured in ‘Framtidens Forskning’

HASTE has been featured in ‘Framtidens Forskning’: “As more and more instruments are generating more and more data, we need new methods to not completely drown in data volumes. Our tools make it possible to know in advance where to focus the analysis, which greatly reduces time-consuming and streamlines resource usage” said Prof. Carolina Wählby, Principle Investigator for the HASTE project. Read the full article.

Publication announcement: Apache Spark Streaming and HarmonicIO: A Performance and Architecture Comparison

The HASTE team are pleased to announce the availability of a new publication of the arXiv pre-print service: ‘Apache Spark Streaming and HarmonicIO: A Performance and Architecture Comparison‘. We performed a benchmark analysis to compare two stream processing frameworks – the popular, Apache Spark framework, widely used in industry, and our own framework HarmonicIO (presented this summer at IEEE Cloud 2018 in San Francisco ).

Previous studies have demonstrated that Apache Spark, Flink and related frameworks can perform stream processing at very high frequencies, but they tend to focus on small messages with a computationally light ‘map’ stage for each message; a common enterprise use case (for example, processing JSON documents). In academic HPC contexts, we often want to analyze larger messages, with more CPU-intensive computations. Our study adds to these benchmarks by broadening the domain to include such processing loads – larger messages (leading to network-bound throughput), and that are computationally intensive (leading to CPU-bound throughput) in the map phase; in order to evaluate applicability of these frameworks to scientific computing applications.

We find that relative performance varies considerably across this domain, with the chosen means of stream source integration having a big impact. Most interestingly, we find that Spark performs very well for large (~10Mb) and small message sizes (~1Kb), but for medium-sized messages, it can be out-performed by HarmonicIO in some configurations. These message sizes are relevant to HASTE, because such file sizes are typical of microscopy applications.

We offer recommendations for choosing and configuring the frameworks, and present a benchmarking toolset developed for this study.

Pre-print is available at: https://arxiv.org/abs/1807.07724

HASTE Meeting in Uppsala & Stockholm

We had a successful project meeting in Uppsala/Stockholm last month – Håkan Wieslander presented his latest research on image feature analysis,  Phil Harrison his latest conformal prediction models, Ben Blamey demonstrated the prototype HASTE pipeline, Niharika Gauraha her work on SVM+. Alan Sabirsh and Johan Karlsson explained a little more about their work at Astrazeneca.

On day 2, we visited Vironova in Stockholm, and were treated to a hands-on demo of their MiniTEM electron microscope – and discussed plans for the next project phase.

Johan preparing a sample.
Demonstrating the MiniTEM
Enjoying Dinner in Uppsala!