Marginalia

Rolf Pfeifer and Josh Bongard argue that intelligence depends on interactions among the body, the brain and the environment.

How the Body Shapes the Way We Think: A New View of Intelligence uses embodied systems to explore that claim.

Robot designs show how physical form and coordination can contribute to behavior.

The discussion connects these ideas with development and the emergence of cognition.

The book offers a way to question accounts of intelligence that focus primarily on computation inside the brain.

Series: AI Books · Book deep dive

Source: How the Body Shapes the Way We Think: A New View of Intelligence · Rolf Pfeifer and Josh Bongard

Explore: AI & Computer Science · AI Books

Related episodes: Neuromorphic Intelligence: Learning, Architectures and Large-Scale Systems (Shuangming Yang & Badong Chen) · Bayesian Models of Cognition: Reverse Engineering the Mind (Thomas Griffiths, Nick Chater & Joshua Tenenbaum)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Any Time, Any Space tackles the problem of keeping generated video shots consistent across changes of place, viewpoint, and time.

The framework turns a screenplay into shot requests with explicit scene, camera, and event-time references.

Scene memory stores registered static geometry, while event memory retains dynamic observations indexed by scene, event, and time.

Before generating a shot, the system retrieves relevant evidence; afterward, review and state checks decide whether new observations enter memory.

The authors report improvements in cross-shot consistency, spatial revisiting, and event replay across four video generators.

Series: Weekly AI Papers · Recent research

Authors: Weijie Wang, Haomin Zhang, Ye Li, Akide Liu, Zeyu Zhang, Haoyu Zhao, Yefei He, Donny Y. Chen, Feng Chen, and Bohan Zhuang

Source: Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation · Weijie Wang, Haomin Zhang, Ye Li et al.

Explore: AI & Computer Science · AI Papers

Related episodes: V-JEPA: Learning Visual Representations from Video (Adrien Bardes, Yann LeCun et al.) · VisionHOPE: Visual Backbones as Self-Modifying Learning Systems

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Kroese and Botev build a rigorous account of probability from measurable spaces, integration, and random variables.

They compare different forms of convergence and develop the law of large numbers and central limit theorem.

Conditional expectation connects information and independence to Markov chains and other stochastic processes.

Martingales, stopping times, and Brownian motion reveal how random systems evolve over time.

The book brings these ideas together through stochastic integration, Itô calculus, and stochastic differential equations.

Series: Mathematics & Science

Book: An Advanced Course in Probability and Stochastic Processes by Dirk P. Kroese and Zdravko I. Botev.

Source: Authors’ official book page · Chapman & Hall/CRC · 2023 · ISBN 9781032320465. Read the free author-provided book.

Explore: More Mathematics & Science episodes · Browse all series

Related episodes: A First Course in Causal Inference (Peng Ding) · Graph Theory (Reinhard Diestel)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Robert Monarch explains how human judgment and machine learning can work together to build better models.

Active learning uses uncertainty and diversity sampling to select examples that most need human feedback.

The book examines how to recruit annotators, write instructions, and measure agreement and annotation quality.

It connects reliable training and evaluation data with transfer learning and other ways to improve labeling workflows.

Annotation interfaces and human-computer interaction bring these methods together in practical machine learning products.

Series: AI Books · Book deep dive

Book: Human-in-the-Loop Machine Learning: Active Learning and Annotation for Human-Centered AI by Robert (Munro) Monarch.

Source: Manning Publications · 2021 · 424 pages · ISBN 9781617296741.

Explore: AI & Computer Science · AI Books · AI Papers

Related episodes: Backpropagation: Learning Representations by Back-Propagating Errors (Rumelhart, Hinton & Williams) · Building Machines That Learn and Think Like People (Brenden Lake et al.)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Dara Horn examines why antisemitism persists and why familiar responses can fail to address it.

The Final Solution to the Jewish Question: A Love Story for the Living combines historical discussion, reporting and personal experience.

Horn questions recurring myths about Jews and examines contemporary responses to anti-Jewish hostility.

She argues for a response that attends to living Jewish communities and their future.

The episode introduces her argument as a contribution to a continuing debate about antisemitism and Jewish life.

Alongside Horn, the conversation draws on economic and historical research, including The Chosen Few by Maristella Botticini and Zvi Eckstein.

Series: History & Judaism

Book: The Final Solution to the Jewish Question: A Love Story for the Living by Dara Horn.

Source: Avid Reader Press/Simon & Schuster · September 1, 2026 · 384 pages · ISBN 9781668090169.

Additional source: The Chosen Few: How Education Shaped Jewish History, 70–1492 (Maristella Botticini & Zvi Eckstein)

Explore: More History & Judaism episodes · Browse all series

Related episodes: Adorno on Antisemitism · Eichmann vs. Arendt

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Backpropagation lets a neural network adjust its internal connections using the errors in its outputs.

David Rumelhart, Geoffrey Hinton and Ronald Williams show how hidden units can learn useful representations in their 1986 Nature paper, Learning representations by back-propagating errors.

A forward pass computes the outputs, and a backward pass calculates how changes in each weight affect the error.

The discussion considers examples in which hidden units discover regularities in the task.

This is a foundational account of an artificial learning procedure, rather than a demonstration that the brain implements it in the same way.

Series: Weekly AI Papers · Historical classic

Authors: David E. Rumelhart, Geoffrey E. Hinton, and Ronald J. Williams

Source: Learning representations by back-propagating errors · David E. Rumelhart, Geoffrey E. Hinton and Ronald J. Williams

Explore: AI & Computer Science · AI Papers

Related episodes: Building Machines That Learn and Think Like People (Brenden Lake et al.) · V-JEPA: Learning Visual Representations from Video (Adrien Bardes, Yann LeCun et al.)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

VisionHOPE explores how a vision model can change both its memory and its learning rules while processing an image.

Its design connects five memories: content, keys, values, learning rate, and retention.

The paper develops controls that keep these coupled updates stable under the conditions established by its analysis.

Four directional scans adapt the method to the two-dimensional structure of images, with computation scaling linearly in the number of image tokens.

The authors report competitive results on ImageNet classification, COCO object detection, and ADE20K semantic segmentation.

This 2026 preprint connects associative memory and Nested Learning to the design of visual backbones.

Series: Weekly AI Papers · Recent research

Authors: Siran Peng, Tianshuo Zhang, Tianyu Fu, Weisong Zhao, Haoyuan Zhang, Jiankuo Zhao, Minghui Wu, Ping Jiang, Xiangyu Zhu, Chenxu Zhao, and Zhen Lei

Source: VisionHOPE: Visual Backbones as Self-Modifying Learning Systems · Siran Peng, Tianshuo Zhang et al.

Explore: AI & Computer Science · AI Papers

Related episodes: V-JEPA: Learning Visual Representations from Video (Adrien Bardes, Yann LeCun et al.) · Building Machines That Learn and Think Like People (Brenden Lake et al.)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Peng Ding introduces causal inference from a statistical perspective.

The book explains the foundations needed to reason about causal effects rather than associations alone.

Examples and applications draw on biostatistics and econometrics.

Mathematical arguments are accompanied by exercises, data, and practical R code.

The course assumes basic probability, statistical inference, and familiarity with regression.

Series: Mathematics & Science · AI Books · Book deep dive

A First Course in Causal Inference — Peng Ding

Publication source

Explore: Mathematics & Science · AI & Computer Science

Related episodes: Graph Theory (Reinhard Diestel) · Lie Groups, Lie Algebras, and Representations (Brian Hall)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Reinhard Diestel’s Graph Theory develops mathematical ways to reason about vertices and the connections between them.

The book studies connectivity, matchings, planarity, coloring and flows.

Proofs explain what can and cannot be inferred from a graph’s structure.

The discussion connects the foundations with more advanced parts of the subject.

This is an introduction to the questions behind a mathematical textbook, with the original work available for the definitions and full proofs.

Series: Mathematics & Science

Graph Theory — Reinhard Diestel

Publication source

Explore: Mathematics & Science

Related episodes: Lie Groups, Lie Algebras, and Representations (Brian Hall) · A First Course in Causal Inference (Peng Ding)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!

Ivan Sag, Thomas Wasow and Emily Bender develop ways to describe which sentence structures a grammar permits.

Syntactic Theory: A Formal Introduction builds analyses through problems that expose the limits of simpler accounts.

Lexical information and constraints help explain how grammatical structures are licensed.

The book connects syntax with meaning and evidence from language processing.

Its exercises invite readers to compare competing analyses and test what each one predicts.

Series: Language & Linguistics · AI Books · Book deep dive

Syntactic Theory: A Formal Introduction — Ivan A. Sag, Thomas Wasow, and Emily M. Bender

Publication source

Explore: Language & Linguistics · AI & Computer Science

Related episodes: Formal Properties of Grammars (Noam Chomsky) · The Sound of Silence: Investigations of Implicit Prosody (Nicholas J. Van Handel)

Got something to say to this podcast? Send a voicemail, and you might hear yourself on the next episode!