My opinions are my own. An introduction to the harm that ML systems cause and to the power imbalance that exists between ML system developers and ML system participants …and 10 concrete ways for machine learning practitioners to help build fairer ML systems. Image description: Photo of Black Lives Matter protesters in Washington, D.C. -- 2 signs say "Black Lives Matter" and "White Silence is Violence." Machine learning systems are increasingly used as tools of oppression. All too often, they're used in high-stakes processes without participants' consent and with no reasonable opportunity for participants to contest the system's decisions -- like when risk assessment systems are used by child welfare services to identify at-risk children; when a machine learning (or "ML") model decides who sees which online ads for employment, housing, or credit opportunities; or when facial recognition systems are used to surveil neighborhoods where Black and Brown people live. In reality though, machine learning systems reflect the beliefs and biases of those who design and develop them.

Researchers at University of Pernambuco in Brazil have come up with a new way of transforming ancient maps in to Google satellite images. Digitisation of maps could inform people of how certain areas have been used and developed over time, including social and economic impacts of urbanisation, the team said in a study titled'Synthesis of Satellite-Like Urban Images From Historical Maps Using Conditional GAN' published in the IEEE GRSL journal. The team used an artificial intelligence (AI) tool called Pix2Pix which is based on two neural networks. The first one creates images based on the input set, while the second network decides if the generated image is fake or not. The networks are trained to fool each other, and ultimately create realistic-looking images based on the historical data provided.

We reduce the task of (span-based) PropBank-style semantic role labeling (SRL) to syntactic dependency parsing. Our approach is motivated by our empirical analysis that shows three common syntactic patterns account for over 98% of the SRL annotations for both English and Chinese data. Based on this observation, we present a conversion scheme that packs SRL annotations into dependency tree representations through joint labels that permit highly accurate recovery back to the original format. This representation allows us to train statistical dependency parsers to tackle SRL and achieve competitive performance with the current state of the art. Our findings show the promise of syntactic dependency trees in encoding semantic role relations within their syntactic domain of locality, and point to potential further integration of syntactic methods into semantic role labeling in the future.

The primary focus of this paper is to produce a proof of concept for extracting drowsiness information from videos to help elderly living on their own. To quantify yawning, eyelid and head movement over time, we extracted 3000 images from captured videos for training and testing of deep learning models integrated with OpenCV library. The achieved classification accuracy for eyelid and mouth open/close status were between 94.3%-97.2%. Visual inspection of head movement from videos with generated 3D coordinate overlays, indicated clear spatiotemporal patterns in collected data (yaw, roll and pitch). Extraction methodology of the drowsiness information as timeseries is applicable to other contexts including support for prior work in privacy-preserving augmented coaching, sport rehabilitation, and integration with big data platform in healthcare.

We consider the problem of sequential graph topology change-point detection from graph signals. We assume that signals on the nodes of the graph are regularized by the underlying graph structure via a graph filtering model, which we then leverage to distill the graph topology change-point detection problem to a subspace detection problem. We demonstrate how prior information on the spectral signature of the post-change graph can be incorporated to implicitly denoise the observed sequential data, thus leading to a natural CUSUM-based algorithm for change-point detection. Numerical experiments illustrate the performance of our proposed approach, particularly underscoring the benefits of (potentially noisy) prior information.

A number of Multiple Criteria Decision Analysis (MCDA) methods have been developed to rank alternatives based on several decision criteria. Usually, MCDA methods deal with the criteria value at the time the decision is made without considering their evolution over time. However, it may be relevant to consider the criteria' time series since providing essential information for decision-making (e.g., an improvement of the criteria). To deal with this issue, we propose a new approach to rank the alternatives based on the criteria time-series features (tendency, variance, etc.). In this novel approach, the data is structured in three dimensions, which require a more complex data structure, as the \textit{tensors}, instead of the classical matrix representation used in MCDA. Consequently, we propose an extension for the TOPSIS method to handle a tensor rather than a matrix. Computational results reveal that it is possible to rank the alternatives from a new perspective by considering meaningful decision-making information.

Maria Zuber got the word on a Friday: Harvard had shut down its research labs. As vice president for research, Zuber consulted with lead researchers across campus over whether MIT should follow suit. "Don't you dare," she remembers them saying. "Don't you dare be like those Harvard people." As covid-19 cases continued to rise across the country, however, she and other senior administrators made the difficult decision by that Sunday: MIT would be scaling down its research to near zero for the first time since opening its doors 155 years ago. "It was a complete shock to people," Zuber says.

Technology is always scrutinized under the lens of scepticism. Despite the many advancements, Artificial Intelligence and its subsidiaries are contributing to; the biases in algorithms remain the biggest challenge amongst experts. Specifically, if the technology is integrated into the draconian laws, the infringement of human rights gets amplified. George Floyd's death casts a shadow on the misuse of technology by authorities. And while tech organizations have apprehended about the negative impact of the technology in society, some organizations are perilously using this technology.

In the latest episode of our podcast, Machine Learning that Works, I had a great pleasure to talk to Gabriel Preda, a Lead Data Scientist at Endava and a Kaggle Grandmaster. For those of you who want to see the full interview, here is the video version. If, on the other hand, you prefer to read, I prepared a summary as well. It's not a faithful transcript of our conversation, but a structured and rephrased version of the interview, that includes the key points and observations. Without further ado, let's meet Gabriel Preda! I work for Endava, which is a software service company, and our projects are actually our clients' projects.

Whether you're logging on from the US, Brazil, Borneo, or France, Facebook can translate virtually any written content published on its platform into the local language using automated machine translation. In fact, Facebook provides around 20 billion translations everyday for its News Feed alone. However these systems typically use English as an intermediary step -- that is, translating from Chinese to French actually goes Chinese to English to French. This is done because data sets of translations to and from English are massive and widely available but putting English in the middle reduces the overall translation accuracy while making the entire process more complex and cumbersome than it needs to be. That's why Facebook AI has developed a new MT model that can bidirectionally translate directly between two languages (Chinese to French and French to Chinese) without ever using English as a crutch -- and which outperforms the English-centric model by 10 points on BLEU metrics.