Cloudy with a high chance of DBMS: a 10-year prediction for enterprise-grade ML

Cloudy with a high chance of DBMS: a 10-year prediction for enterprise-grade ML, Agrawal et al., CIDR'20 "Cloudy with a high chance of DBMS" is a fascinating vision paper from a group of experts at Microsoft, looking at the transition of machine learning from being primarily the domain of large-scale, high-volume consumer applications to being ... Continue Reading

Programmatically interpretable reinforcement learning

Programmatically interpretable reinforcement learning, Verma et al., ICML 2018 Being able to trust (interpret, verify) a controller learned through reinforcement learning (RL) is one of the key challenges for real-world deployments of RL that we looked at earlier this week. It's also an essential requirement for agents in human-machine collaborations (i.e, all deployments at some ... Continue Reading

Challenges of real-world reinforcement learning

Challenges of real-world reinforcement learning, Dulac-Arnold et al., ICML'19 Last week we looked at some of the challenges inherent in automation and in building systems where humans and software agents collaborate. When we start talking about agents, policies, and modelling the environment, my thoughts naturally turn to reinforcement learning (RL). Today's paper choice sets out ... Continue Reading

Optimized risk scores

Optimized risk scores Ustun & Rudin, KDD'17 On Monday we looked at the case for interpretable models, and in Wednesday’s edition of The Morning Paper we looked at CORELS which produces provably optimal rule lists for categorical assessments. Today we’ll be looking at RiskSLIM, which produces risk score models together with a proof of optimality. ... Continue Reading

Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead

Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead Rudin et al., arXiv 2019 With thanks to Glyn Normington for pointing out this paper to me. It’s pretty clear from the title alone what Cynthia Rudin would like us to do! The paper is a mix of technical ... Continue Reading