S++: A Fast and Deployable Secure-Computation Framework for Privacy-Preserving Neural Network Training

31 de January de 2021 31 de January de 2021

We introduce S++, a simple, robust, and deployable framework for training a neural network (NN) using private data from multiple sources, using secret-shared secure function evaluation. In short, consider a virtual third party to whom every data-holder sends their inputs, and which computes the neural network: in our case, this virtual third party is actually a set of servers which individually learn nothing, even with a malicious (but non-colluding) adversary.

Pages: 1 2 3

Multi-Image Steganography Using Deep Neural Networks

23 de January de 2021 31 de January de 2021

Steganography is the science of hiding a secret message within an ordinary public message. Over the years, steganography has been used to encode a lower resolution image into a higher resolution image by simple methods like LSB manipulation. We aim to utilize deep neural networks for the encoding and decoding of multiple secret images inside a single cover image of the same resolution.

Pages: 1 2 3

Addressing Ethical Dilemmas in AI: Listening to Engineers

16 de January de 2021 22 de August de 2021

Documentation is key – design decisions in AI development must be documented in detail, potentially taking inspiration from the field of risk management. There is a need to develop a framework for large-scale testing of AI effects, beginning with public tests of AI systems, and moving towards real-time validation and monitoring. Governance frameworks for decisions in AI development need to be clarified, including the questions of post-market surveillance of product or system performance. Certification of AI ethics expertise would be helpful to support professionalism in AI development teams. Distributed responsibility should be a goal, resulting in a clear definition of roles and responsibilities as well as clear incentive structures for taking in to account broader ethical concerns in the development of AI systems.

Pages: 1 2 3

Explainability in Graph Neural Networks: A Taxonomic Survey

7 de January de 2021 7 de January de 2021

We summarize current datasets and metrics for evaluating GNN explainability. Altogether, this work provides a unified methodological treatment of GNN explainability and a standardized testbed for evaluations.

Pages: 1 2 3

R0:daa46e0033d013818bcc056d269d5737-Unsupervised deep clustering and reinforcement learning can accurately segment MRI brain tumors with very small training sets

Unsupervised deep clustering and reinforcement learning can accurately segment MRI brain tumors with very small training sets

2 de January de 2021 2 de January de 2021

“We have demonstrated a proof-of-principle application of unsupervised deep clustering and reinforcement learning to segment brain tumors. The approach represents human-allied AI that requires minimal input from the radiologist without the need for hand-traced annotation”.

Pages: 1 2 3

Fourier Neural Operator for Parametric Partial Differential Equations

29 de October de 2020 11 de November de 2020

The classical development of neural networks has primarily focused on learning mappings between finite-dimensional Euclidean spaces. Recently, this has been generalized to neural operators that learn mappings between function spaces. For partial differential equations (PDEs), neural operators directly learn the mapping from any functional parametric dependence to the solution.

Pages: 1 2 3

Evaluating and Characterizing Human Rationales

20 de October de 2020 11 de November de 2020

Two main approaches for evaluating the quality of machine-generated rationales are: 1) using human rationales as a gold standard; and 2) automated metrics based on how rationales affect model behavior.

Pages: 1 2 3

https://editorialia.com/2020/10/11/publications-r0identifier_7c166ecb19775d64f5f8a16fce100ea9-medical-notes-summariser-characterizing-the-value-of-information-in-medical-notes

Medical notes summariser: “Characterizing the Value of Information in Medical Notes”

11 de October de 2020 11 de November de 2020

Machine learning models depend on the quality of input data. As electronic health records are widely adopted, the amount of data in health care is growing, along with complaints about the quality of medical notes.

Pages: 1 2 3

->Artificial intelligence towards data science

Probabilistic Machine Learning for Healthcare

27 de September de 2020 9 de October de 2020

Machine learning can be used to make sense of healthcare data. Probabilistic machine learning models help provide a complete picture of observed data in healthcare. In this review, we examine how probabilistic machine learning can advance healthcare. We consider challenges in the predictive model building pipeline where probabilistic models can be beneficial including calibration and missing data. Beyond predictive models, we also investigate the utility of probabilistic machine learning models in phenotyping, in generative models for clinical use cases, and in reinforcement learning.

Pages: 1 2 3

Report on Publications Norms for Responsible AI

20 de September de 2020 20 de September de 2020

The history of science and technology shows that seemingly innocuous developments in scientific theories and research have enabled real-world applications with significant negative consequences for humanity.

Pages: 1 2 3

English (Paper)