The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

Amir Zamir, an Assistant Professor of Computer Science at the Swiss Federal Institute of Technology, dives into the exciting advancements in computer vision. He discusses how the field has evolved, particularly in 3D vision and self-supervised learning, which reduces reliance on labeled data. The conversation touches on the challenges of navigating unseen spaces for robotics and the significance of multitask learning for improving network robustness. Zamir also explores the practical applications of these technologies, including their potential for autonomous driving and real-world problem-solving.

In this engaging discussion, Nasrin Mostafazadeh, a Senior AI Research Scientist at Elemental Cognition, shares her insights on the evolution of Natural Language Processing (NLP). She highlights the transformative impact of large pre-trained models like BERT and GPT-2. Nasrin dives into the ethical implications of AI, including bias and accessibility, and stresses the importance of interpretability in AI systems. The conversation also touches on the challenges of AI in educational assessments and aims to enhance common sense reasoning within NLP.

Timnit Gebru, a research scientist at Google Brain and co-lead of their ethical AI team, dives into the evolving landscape of AI fairness and ethics. She discusses the importance of representation, highlighting initiatives like Black in AI that enhance diversity in tech conferences. Gebru also reflects on the significance of intersectional testing and the introduction of model cards for transparency. With insights from NeurIPS, she navigates the challenges of promoting ethical AI amidst personal and institutional hurdles, underscoring the need for inclusive voices in shaping AI governance.

Chelsea Finn, Assistant Professor at Stanford University, shares her insights on advancements in reinforcement learning. She breaks down model-based approaches and the challenges of exploration in complex environments like Montezuma's Revenge. The discussion also touches on the importance of curriculum learning in robotics and the nuances of batch off-policy learning. With exciting implications for real-world applications, Chelsea highlights the evolving landscape of RL libraries and their role in bridging the gap between simulation and practical deployment.

In this engaging discussion, Zack Lipton, a Professor at CMU with expertise in machine learning, explores key advances from 2019 in the field. He delves into the evolution of deep learning, noting the impact of models like BERT and challenges related to distribution shifts. Lipton also discusses innovative approaches in causal inference and fairness, advocating for continued research on model robustness. Lastly, he shares predictions about commodification in AI and the need for inclusive participation in the future landscape of machine learning.

Join Mohamed Sidahmed, R&D Manager at Shell, as he discusses groundbreaking advancements in machine learning and AI at NeurIPS. He dives into the innovative FaciesNet architecture, which transforms geological data into spectrograms for improved rock facies classification. Learn how these techniques revolutionize seismic imaging and enhance predictive capabilities, ultimately boosting hydrocarbon exploration confidence. Sidahmed also highlights the vital role of collaboration between academia and industry in driving energy-related AI innovations.

Daphne Koller, the co-founder of Coursera and CEO of Insitro, shares her expertise on the revolutionary role of machine learning in drug discovery. She discusses the hurdles of the pharmaceutical landscape, including high costs and regulatory challenges. Koller emphasizes how ML can streamline decision-making and enhance drug efficacy through targeted therapies. Highlighting innovative techniques like CRISPR and high throughput biology, she stresses the need for collaboration between biology and tech experts to transform healthcare.

Blake Richards, Assistant Professor at McGill University and Core Faculty Member at Mila, dives into the brain's learning abilities with a focus on sensory prediction error signals. He elaborates on two-photon calcium imaging studies revealing how the neocortex processes unexpected stimuli. Discussing predictive coding, he highlights its implications for both neuroscience and machine learning. The conversation also touches on integrating memory systems in reinforcement learning, showcasing how insights from biology can lead to more adaptive AI.

In a captivating discussion, Celeste Kidd, an Assistant Professor of Psychology at UC Berkeley, explores how we form beliefs and our curiosity about the world. She explains the role of past experiences in shaping future interests and how certainty can lead to rigidity in thought. The conversation also delves into the interplay between attention, decision-making, and how infants develop probabilistic expectations. Kidd's insights reveal the complexities of knowledge acquisition and the implications for both individuals and intelligent systems.

Feng Yan, an Assistant Professor at the University of Nevada, Reno, is at the forefront of using machine learning for wildfire prediction. He introduces ALERTWildfire, a network of cameras that capture real-time data to enhance monitoring efforts. The conversation dives into innovative camera deployments, the integration of satellite and ground-level data, and overcoming challenges in model training. Feng also discusses leveraging IaaS and FaaS for scalability and cost-effectiveness in tackling the growing threat of wildfires.