Step 1. Place your order
Fill in the order form and provide all details of your assignment.
Step 2. Make Payment
Choose the payment system that suits you most.
Step 3. Receive your paper
TOPIC: Analyse how the performance of ML algorithms being affected by geographic
Get Help With Your Essay
"Place your order now for a similar assignment and have exceptional work written by our team of experts, guaranteeing you A results."
TOPIC: Analyse how the performance of ML algorithms being affected by geographical region and/or year
I would suggest for you to look at WILDS is a curated collection of benchmark datasets that represent distribution shifts faced in the wild (https://wilds.stanford.edu/datasets/). Particularly, a dataset called FMoW (https://wilds.stanford.edu/datasets/#motivation-6).
Motivation – FMoW
ML models for satellite imagery can enable global-scale monitoring of sustainability and economic challenges, aiding policy and humanitarian efforts in applications such as tracking deforestation (Hansen et al., 2013), population density mapping (Tiecke et al., 2017), crop yield prediction (Wang et al., 2020b), and other economic tracking applications (Katona et al., 2018). As satellite data constantly changes due to human activity and environmental processes, these models must be robust to distribution shifts over time. Moreover, as there can be disparities in the data available between regions, these models should ideally have uniformly-high accuracies instead of only doing well on data-rich regions and countries.
We study this problem on a variant of the Functional Map of the World dataset (Christie et al., 2018).
Problem setting – FMoW
We consider a hybrid domain generalization and subpopulation shift problem, where the input x is a RGB satellite image (resized to 224 x 224 pixels), the label y is one of 62 building or land use categories, and the domain d represents both the year the image was taken as well as its geographical region (Africa, the Americas, Oceania, Asia, or Europe). We aim to solve both a domain generalization problem across time and improve subpopulation performance across regions.
The dataset and code for a lot of methods is available at https://github.com/p-lambda/wilds/. You can try to analyse how the performance of ML algorithms being affected by geographical region and/or year.
Dissertations should demonstrate a mix of skills at masters level. Depending on the project, different skills will have greater weighting in the marking, but no project will be carried by a single one. The criteria for evaluation include:
application of or extension of MSc course skills, ideally beyond those taught in that course, or skills beyond those that might reasonably be expected of a computer science undergraduate.
engagement with the literature, including appropriate selection of papers and analysis of concepts.A dissertation which applies concepts from one field in another area or combines concepts from two fields may attract greater weighting for the literature aspect.
theoretical analysis and development of concepts.
quality of programming, proofs and other practical development work. Organisation, clarity, efficiency, application of advanced methods and novelty are the focus. A large volume of code is not, by itself, sufficient.
quality of evaluation, including choice of methods, controls and conditions; rigour of their application, and analysis of data.
novelty is not an absolute requirement of an MSc dissertation. However, the work undertaken should engage with recent developments in computer science. Where there is novelty, for instance modification of algorithms or a new approach to a proof, this shall be acknowledged in the marking.
good project management will be reflected in the outcomes of the dissertation. However, examiners may wish to note appropriate selection of tools and methods and suitable management of time and risk, particularly where engaging with very new tools or making novel contributions.
General professional standards will be expected:
in matters of punctuation, vocabulary choice, standard English grammar, and the conventions of academic discourse (including reference to sources).
in presentation of code (for programming projects): Code for programming projects should be submitted as an appendix to the main report.
in formal aspects of presentation (word-processing/typing, printing).
Guidelines to students and markers on standards expected at each level
70% – 100% – Excellent
Shows very good understanding supported by evidence that the student has gone beyond what was taught by extra study, programming, or creative thought. Work at the top end of this range is of exceptional quality.
Write-up: well-structured, correct references, critical discussion of existing relevant work, neatly presented, interesting and clearly expressed, thorough disinterested critique of what is good and bad about the approach taken, and proposals about how the project work could be developed in the future.
Literature: engagement with current research, including appropriate analysis, comparison, critique and selection and precis of key ideas relating to the student’s work.
Program: code that executes efficiently, incorporates sophisticated programming features, is non-redundant, well-structured, well commented and elegant, addresses the problem effectively for a non-trivial application.
Theoretical analysis: appropriate application of techniques, including classification, proof, complexity analysis etc., to a non-trivial problem.
Evaluation: a substantial evaluation, through appropriate interpretation of analysis, simulation, deployment, functional and non-functional testing, etc., coupled with excellent interpretation and presentation of results. In more experimental projects results will be repeatable and contain comparison with alternative techniques and/or significant exploration of the parameters of the code/problem.
60% – 69% – Good
Very competent in all respects, substantially correct and complete knowledge but not going beyond what was taught.
Literature: engagement with literature, including critique of ideas, well-related to the rest of the project. Possibly not engaging beyond further reading from course(s).
Program: code that executes, incorporates some complexity, is well-designed and presented and addresses a reasonably non-trivial problem related to the literature.
Theoretical analysis: a clear, if not particularly sophisticated, analysis of the problem.
Evaluation: a good application of appropriate techniques leading to a clear result.
55% – 59% – Satisfactory
Competent in most respects. Minor gaps in knowledge but reasonable understanding of fundamental concepts.
Literature: a presentation of ideas from the literature, given some structure and basic analysis and related to the rest of the project.
Program: code that executes, and addresses a simple problem.
Theoretical analysis: a competent analysis of at least the most significant concepts in the work.
Evaluation: evidence of appropriate testing beyond function testing of code.
50% – 54% – Borderline
Significant gaps in knowledge but some understanding of funamental concepts. Typically this project will be a marginal development or integration of course or textbook ideas, with some evaluation or analysis.
0% – 49% – Fail
Inadequate knowledge of the subject. Work is seriously flawed, displaying major lack of understanding, irrelevance or incoherence. Code, analysis and evaluation that are not coherent in terms of the problem being addressed or the methods to be employed in doing this.
Only pay once! Clear pricing system with no hidden fees, direct line to your writer, 24/7 customer support.
Our reputation is built on delivering top-quality papers. No matter how small the task, we give customers our best.
Missed due dates can result in a serious loss of marks. We strive to finish everything with time to spare.