ATGO AI | Accountability, Trust, Governance and Oversight of Artificial Intelligence |

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems. These are published as podcast series.

In this podcast we have Kathrin Grosse. Kathrin Grosse is a Post Doc researcher with Battista Biggio at the University of Cagliari working on Adversarial learning.

This podcast covers a paper titled “Machine Learning Security against Data Poisoning: Are We There Yet? ” published in April 2022, which she co-authored.

This is part 2 of the podcast. In this podcast, she covers the thoughts around gaining a better understanding of how defenses work, adaptive attacks and thus, our knowledge about the limits of existing defenses is rather narrow

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems. These are published as podcast series.

In this podcast we have Kathrin Grosse. Kathrin Grosse is a Post Doc researcher with Battista Biggio at the University of Cagliari working on Adversarial learning.

In this podcast we cover a paper titled “Machine Learning Security against Data Poisoning: Are We There Yet? ” published in April 2022, which she co-authored.

This is part 1 of the podcast. In this podcast, she covers the thoughts around the impracticality of some threat models considered for poisoning attacks in a real-world application and scalability of poisoning attacks against large-scale models

—

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems. These are published as podcast series.

Ideas emerge when curiosity meets clarity. Here we go with OPENBOX to bring clarity to those curious minds looking to solve real-world problems. This project is done in collaboration with ForHumanity. ForHumanity is a 501(c)(3) nonprofit organization dedicated to minimizing the downside risks of AI and autonomous systems. ForHumanity develops criteria for an independent audit of AI systems. To know more, visit https://forhumanity.center/. 

Today, we have Raghu with us. Raghu is a Ph.D. student at the Machine Learning Group at the Univerity of Freiburg, under the supervision of Frank Hutter. He is working on automating hyperparameter optimization for RL, AutoRL. His master's thesis was on Reinforcement learning. Artificial General Intelligence is an area of interest to him in the long term. He is also exploring Dynamic Algorithm configuration (Controlling hyperparameter dynamically). We will cover a paper titled “Automated Reinforcement Learning (AutoRL): A Survey and Open Problems” published in June 2022, which he co-authored. 

This is part 3 of the discussion. In this part, he covers the open issues in hyper parameter optimization using the Environmental design, Hybrid approaches and Benchmarks.

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems. These are published as podcast series.

Ideas emerge when curiosity meets clarity. Here we go with OPENBOX to bring clarity to those curious minds looking to solve real-world problems. 

This project is done in collaboration with ForHumanity. ForHumanity is a 501(c)(3) nonprofit organization dedicated to minimizing the downside risks of AI and autonomous systems. ForHumanity develops criteria for an independent audit of AI systems. To know more, visit https://forhumanity.center/. 

Today, we have Raghu with us. Raghu is a Ph.D. student at the Machine Learning Group at the Univerity of Freiburg, under the supervision of Frank Hutter. He is working on automating hyperparameter optimization for RL, AutoRL. His master's thesis was on Reinforcement learning. Artificial General Intelligence is an area of interest to him in the long term. He is also exploring Dynamic Algorithm configuration (Controlling hyperparameter dynamically). We will cover a paper titled “Automated Reinforcement Learning (AutoRL): A Survey and Open Problems” published in June 2022, which he co-authored. 

This is part 2 of the discussion. In this part, he covers the open issues in Evolutionary approaches, Meta gradient for online tuning and Blackbox online tuning.

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems. These are published as podcast series.

Ideas emerge when curiosity meets clarity. Here we go with OPENBOX to bring clarity to those curious minds looking to solve real-world problems.

This project is done in collaboration with ForHumanity. ForHumanity is a 501(c)(3) nonprofit organization dedicated to minimizing the downside risks of AI and autonomous systems. ForHumanity develops criteria for an independent audit of AI systems. To know more, visit https://forhumanity.center/.

Today, we have Raghu with us. Raghu is a Ph.D. student at the Machine Learning Group at the Univerity of Freiburg, under the supervision of Frank Hutter. He is working on automating hyperparameter optimization for RL, AutoRL.

His master's thesis was on Reinforcement learning. Artificial General Intelligence is an area of interest to him in the long term. He is also exploring Dynamic Algorithm configuration (Controlling hyperparameter dynamically).

We will cover a paper titled “Automated Reinforcement Learning (AutoRL): A Survey and Open Problems” published in June 2022, which he co-authored.

This is part 1 of the discussion. In this part, he covers the open issues in hyper parameter optimization using the Random grid search approach and Bayesian optimization. 

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems.

Today, we have with us paul. Paul is a PhD Student at Barcelona Neural Networking Center Technical University of Catalunya working on the use of ML to solve problems in communication networks. We are going to cover a paper titled “Towards Real-Time Routing Optimization with Deep Reinforcement Learning: Open Challenges ” published recently which he co-authored. In this podcast, he is covering aspects of (a) Training time and cost associated with Deep Reinforcement Learning and (b) lack of performance bounds. This is part 2 of the podcast

This project is in collaboration with ForHumanity. ForHumanity is a 501(c)(3) nonprofit organization with a mission to minimize the downside risks of AI and autonomous systems. ForHumanity develops criteria for an independent audit of AI systems. To know more, visit https://forhumanity.center/.

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems.

Today, we have with us paul. Paul is a PhD Student at Barcelona Neural Networking Center Technical University of Catalunya working on the use of ML to solve problems in communication networks. We are going to cover a paper titled “Towards Real-Time Routing Optimization with Deep Reinforcement Learning: Open Challenges ” published recently which he co-authored. In this podcast, he is covering aspects of (a) Generalization in Deep Reinforcement Learning and (b) Defining an appropriate action space. This is part 1 of the podcast

This project is in collaboration with ForHumanity. ForHumanity is a 501(c)(3) nonprofit organization with a mission to minimize the downside risks of AI and autonomous systems. ForHumanity develops criteria for an independent audit of AI systems. To know more, visit https://forhumanity.center/.

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems.

In this episode, Rafael Figueiredo Prudencio discusses open issues in Offline Reinforcement Learning. He covers aspects relating to (a) Approximation function and generalization and (b) leveraging unlabelled data. Conversation with Rafael is 2 part podcast series, and this podcast is part 2. Listen to the podcast to understand specific ethical issues arising from the open issues.

This project is in collaboration with ForHumanity. ForHumanity is a 501(c)(3) nonprofit organization with a mission to minimize the downside risks of AI and autonomous systems. ForHumanity develops criteria for an independent audit of AI systems. To know more, visit https://forhumanity.center/.

OPENBOX aims at bringing an easier understanding of open problems that helps in finding solutions for such problems. For the said purpose, I interview researchers and practitioners who have published works on open problems in various areas of Artificial Intelligence and Machine Learning to collect a simplified understanding of these open problems. 

In this episode, Rafael Figueiredo Prudencio discusses open issues in Offline Reinforcement Learning. He covers aspects relating to (a) Off Policy evaluation methods and (b) the lack of adequate real-world benchmarks. Conversation with Rafael is 2 part podcast series, and this podcast is part 1. Listen to the podcast to understand specific ethical issues arising from the open issues.

This project is in collaboration with ForHumanity. ForHumanity is a 501(c)(3) nonprofit organization with a mission to minimize the downside risks of AI and autonomous systems. ForHumanity develops criteria for an independent audit of AI systems. To know more, visit https://forhumanity.center/.

ATGO AI is a podcast channel from ForHumanity. This podcast will bring multiple series of insights on topics of pressing importance specifically in the space of Ethics and Accountability of emerging technology. You will hear from game changers in this field who have spearheaded accountability, transparency, governance and oversight in developing and deploying emerging technology (including Artificial Intelligence).

Joshua Bucheli is an international polyglot. He is based in Switzerland and has a background in event and project management as well as applied ethical research and critical-analysis. He is an independent researcher, writer, and editor with an MA in Political, Legal, and Economic Philosophy and has experience ranging from humanitarian fieldwork for the UNHCR in Malaysia to researching and editing for academic journal articles and Swiss think tanks on the subject of ethical AI and robotics. He is a former Head of Community Management at Let’s Phi and a freelance writer for cyberunity on the subject of careers in cybersecurity.

Joshua shares that he is working on a code of ethics in order to build an industry standard for ethics and AI. The draft EU AI regulations will help set a course - will there be a more human centered focus in regards to technological innovation? Taking a careful approach towards the ethical grey area is impactful. Pause is needed to evaluate and mitigate potential harm.

Visit us at https://forhumanity.center/ to learn more