Federated Learning - Revision history

Tom: /* References */

2025-05-28T23:40:33Z

‎References

Tom: /* Key Deliverables */

2024-01-12T00:38:19Z

‎Key Deliverables

Tom: /* Definitions */

2024-01-12T00:24:41Z

‎Definitions

Tom: /* Privacy Enhancing */

2022-09-12T22:46:17Z

‎Privacy Enhancing

Tom: /* Definitions */

2022-09-09T23:38:18Z

‎Definitions

Tom: /* Definitions */

2022-09-09T23:37:26Z

‎Definitions

Tom: /* Privacy Enhancing */

2022-09-07T04:37:23Z

‎Privacy Enhancing

Tom: /* Privacy Enhancing */

2022-09-07T04:36:46Z

‎Privacy Enhancing

Tom: /* Privacy Enhancing */

2022-09-07T04:32:30Z

‎Privacy Enhancing

Tom: /* Privacy Enhancing */

2022-09-07T04:31:33Z

‎Privacy Enhancing

@@ Line 63: / Line 63: @@
 <references />
+[[Category: Learning]]
 [[Category: Artificial Intelligence]]
 [[Category: Privacy]]

@@ Line 40: / Line 40: @@
 ==Key Deliverables==
-# An interface to the leaf nodes that accepts requests for new learning and passes back learning into multiple aggregators.
+# An interface to the [[Leaf Node]]s that accepts requests for new learning and passes back learning into multiple aggregators.
-# A means to motivate the leaf nodes to participate in in a variety of research goals.
+# A means to motivate the [[Leaf Node]]s to participate in in a variety of research goals.
 ===Intelligent Goals===
-The objective the effort is a description of the results that can be passed from the leaf nodes to the final analysis. Only recently has AI begun to address this as described in the wiki page [[Intelligent Goals]].
+The objective the effort is a description of the results that can be passed from the [[Leaf Node]]s to the final analysis. Only recently has AI begun to address this as described in the wiki page [[Intelligent Goals]].
 ==Using Test Data==

@@ Line 15: / Line 15: @@
 '''Federated learning''' is a machine learning setting where multiple entities (clients) collaborate in solving a machine learning problem, under the coordination of a central server or service provider. Each client's raw data is stored locally and not exchanged or transferred; instead, focused updates intended for immediate aggregation are used to achieve the learning objective. <ref>P. Kairouz, et al., ''Advances and open problems in federated learning'' Foundations and Trends in Machine Learning: 14 (1-2); (2021) https://arxiv.org/abs/1912.04977.</ref>
-'''Decentralized learning''' might be a better fit definition for this work as the is NO SINGLE CENTRAL AUTHORITY. Rather the leaf nodes are structured to accept requests from a variety of sources as well as their own local learning suited to their own local requests. Note that this could also be viewed as learning with more than one federation.
+'''Decentralized learning''' might be a better fit definition for this work as the is NO SINGLE CENTRAL AUTHORITY. Rather the [[Leaf Node]]s are structured to accept requests from a variety of sources as well as their own local learning suited to their own local requests. Note that this could also be viewed as learning with more than one federation.
 ==A Hierarchical Directed Graph==

@@ Line 31: / Line 31: @@
 ==Privacy Enhancing==
 *Three key components of privacy are used in this context:<ref>Kallista Bonawitz +3, (Google) ''Federated Learning and Privacy'' ACM Queue (2021-11-16) https://queue.acm.org/detail.cfm?id=3501293</ref>
-# user understanding and consent;
+# User understanding and consent;
-# data minimization (collect only the data needed for the specific computation); and
+# [[Data Minimization]] (collect only the data needed for the specific computation); and
-# anonymization (the final released output of the computation does not reveal anything unique to an individual) of released aggregates.
+# Anonymization (the final released output of the computation does not reveal anything unique to an individual) of released aggregates.
 * An [https://ktn-uk.org/wp-content/uploads/2022/08/PETs-Prize-Challenges_-Public-Health-Technical-Brief-1.pdfinitiative of the US and UK governments], Privacy Enhancing Technologies, has [https://www.drivendata.org/competitions/98/nist-federated-learning-1/page/525/ one track focused on Transforming Pandemic Response and Forecasting through Federated Learning with End-to-End Privacy].
 * To make the graph privacy-enhancing we demand that any personally identifiable information (PII) is restricted to the leaves meaning that it never is processed anywhere but within the medical facility where it is already maintained with full personal data. Only the learning from the data is passed up to the aggregator.

@@ Line 15: / Line 15: @@
 '''Federated learning''' is a machine learning setting where multiple entities (clients) collaborate in solving a machine learning problem, under the coordination of a central server or service provider. Each client's raw data is stored locally and not exchanged or transferred; instead, focused updates intended for immediate aggregation are used to achieve the learning objective. <ref>P. Kairouz, et al., ''Advances and open problems in federated learning'' Foundations and Trends in Machine Learning: 14 (1-2); (2021) https://arxiv.org/abs/1912.04977.</ref>
-'''Decentralized learning''' might be a better fit definition for this work as the is NO SINGLE CENTRAL AUTHORITY. Rather the leaf nodes are structured to accept requests from a variety of sources as well as their own local learning suited to their own local requests.
+'''Decentralized learning''' might be a better fit definition for this work as the is NO SINGLE CENTRAL AUTHORITY. Rather the leaf nodes are structured to accept requests from a variety of sources as well as their own local learning suited to their own local requests. Note that this could also be viewed as learning with more than one federation.
 ==A Hierarchical Directed Graph==

@@ Line 37: / Line 37: @@
 * To make the graph privacy-enhancing we demand that any personally identifiable information (PII) is restricted to the leaves meaning that it never is processed anywhere but within the medical facility where it is already maintained with full personal data. Only the learning from the data is passed up to the aggregator.
 * The question to be determine is where this privacy guarantee will adversely impact the accuracy of the learning process. The test matrix proposed here will address that.
-* Another question might be if a small hospital in some impoverished state or country delivered its learning to show particularly poor results. Should we consider that learning to be data leakage on the residents of that hospital? Perhaps those impoverished regions would not want their data to be known outside of the broader learning result?<ref>Jiale Chen + 4''Beyond Model-Level Membership Privacy Leakage: an Adversarial Approach in Federated Learning'' IEEE 2020 Int'l conference https://ieeexplore.ieee.org/abstract/document/9209744</ref>
+* Another question might be if a small hospital in some impoverished state or country delivered its learning to show particularly poor results. Should we consider that learning to be data leakage on the residents of that hospital? Perhaps those impoverished regions would not want their data to be known outside of the broader learning result?<ref>Jiale Chen + 4, ''Beyond Model-Level Membership Privacy Leakage: an Adversarial Approach in Federated Learning'' IEEE 2020 Int'l conference https://ieeexplore.ieee.org/abstract/document/9209744</ref>
 ==Key Deliverables==