ANC Workshop - 05/11/2019

Host: Amos Storkey

 

Speaker: Michael Camilleri

Title: The Extended Dawid-Skene Model: Fusing Information from Multiple Data Schemas

Abstract:  While label fusion from multiple noisy annotations is a well understood concept in data wrangling (tackled for example by the Dawid-Skene (DS) model), we consider the extended problem of carrying out learning when the labels themselves are not consistently annotated with the same schema. We show that even if annotators use disparate, albeit related, label-sets, we can still draw inferences for the underlying full label-set. We propose the Inter-Schema AdapteR (ISAR) to translate the fully-specified label-set to the one used by each annotator, enabling learning under such heterogeneous schemas, without the need to re-annotate the data. We apply our method to a mouse behavioural dataset, achieving significant gains (compared with DS) in out-of-sample log-likelihood (-3.40 to -2.39) and F1-score (0.785 to 0.864).

 

Speaker: Emily Scher

Title:  Human-Genome Interaction: The Gap Between How DNA is Designed, and How It Functions In Vivo

AbstractSynthetic Biology – the field in which Biologists design DNA sequences and implement them in vivo – often ignores the epigenetic context around an inserted DNA sequence. However, as the field has taken on more and more complex experiments in higher order organisms, the gap between how DNA is viewed in the field and how it exists in reality has widened. This talk will introduce a number of projects under the umbrella of bridging this gap, focusing on an attempt to codify some aspects of epigenetics using various models to resolve the grammatical patterns which govern their occupancy.

Nov 05 2019 -

ANC Workshop - 05/11/2019

The Extended Dawid-Skene Model: Fusing Information from Multiple Data Schemas

G.03, IF