Hattie Zhou: Lottery Tickets and Algorithmic Reasoning in LLMs

Hattie Zhou: Lottery Tickets and Algorithmic Reasoning in LLMs

Author: Daniel Bashir February 16, 2023 Duration: 1:42:59

In episode 60 of The Gradient Podcast, Daniel Bashir speaks to Hattie Zhou.

Hattie is a PhD student at the Université de Montréal and Mila. Her research focuses on understanding how and why neural networks work, based on the belief that the performance of modern neural networks exceeds our understanding and that building more capable and trustworthy models requires bridging this gap. Prior to Mila, she spent time as a data scientist at Uber and did research with Uber AI Labs.

Have suggestions for future podcast guests (or other feedback)? Let us know here!

Subscribe to The Gradient Podcast:  Apple Podcasts  | Spotify | Pocket Casts | RSSFollow The Gradient on Twitter

Outline:

* (00:00) Intro

* (01:55) Hattie’s Origin Story, Uber AI Labs, empirical theory and other sorts of research

* (10:00) Intro to the Lottery Ticket Hypothesis & Deconstructing Lottery Tickets

* (14:30) Lottery tickets as lucky initialization

* (17:00) Types of masking and the “masking is training” claim

* (24:00) Type-0 masks and weight evolution over long training trajectories

* (27:00) Can you identify good masks or training trajectories a priori?

* (29:00) The role of signs in neural net initialization

* (35:27) The Supermask

* (41:00) Masks to probe pretrained models and model steerability

* (47:40) Fortuitous Forgetting in Connectionist Networks

* (54:00) Relationships to other work (double descent, grokking, etc.)

* (1:01:00) The iterative training process in fortuitous forgetting, scale and value of exploring alternatives

* (1:03:35) In-Context Learning and Teaching Algorithmic Reasoning

* (1:09:00) Learning + algorithmic reasoning, prompting strategy

* (1:13:50) What’s happening with in-context learning?

* (1:14:00) Induction heads

* (1:17:00) ICL and gradient descent

* (1:22:00) Algorithmic prompting vs discovery

* (1:24:45) Future directions for algorithmic prompting

* (1:26:30) Interesting work from NeurIPS 2022

* (1:28:20) Hattie’s perspective on scientific questions people pay attention to, underrated problems

* (1:34:30) Hattie’s perspective on ML publishing culture

* (1:42:12) Outro

Links:

* Hattie’s homepage and Twitter

* Papers

* Deconstructing Lottery Tickets: Zeros, signs, and the Supermask

* Fortuitous Forgetting in Connectionist Networks

* Teaching Algorithmic Reasoning via In-context Learning



Get full access to The Gradient at thegradientpub.substack.com/subscribe

Hosted by Daniel Bashir, The Gradient: Perspectives on AI moves beyond surface-level headlines to explore the intricate machinery and human ideas shaping artificial intelligence. Each episode is built on a foundation of deep research, leading to conversations that are both technically substantive and broadly accessible. You'll hear from researchers, engineers, and philosophers who are actively building and critiquing our technological future, discussing not just how AI systems work, but the larger implications of their integration into society. This isn't about speculative hype; it's a grounded examination of real progress, persistent challenges, and ethical considerations from those on the front lines. The discussions peel back layers on topics like model architecture, policy, and the fundamental science behind the algorithms becoming part of our daily lives. For anyone curious about the substance behind the buzz-whether you have a technical background or are simply keen to understand a defining technology of our age-this podcast offers a crucial and thoughtful resource. Tune in for a consistently detailed and nuanced take that treats artificial intelligence with the complexity it deserves.
Author: Language: English Episodes: 100

The Gradient: Perspectives on AI
Podcast Episodes
2025 in AI, with Nathan Benaich [not-audio_url] [/not-audio_url]

Duration: 1:01:15
Episode 144Happy New Year! This is one of my favorite episodes of the year — for the fourth time, Nathan Benaich and I did our yearly roundup of AI news and advancements, including selections from this year’s State of AI…
Iason Gabriel: Value Alignment and the Ethics of Advanced AI Systems [not-audio_url] [/not-audio_url]

Duration: 58:39
Episode 143I spoke with Iason Gabriel about:* Value alignment* Technology and worldmaking* How AI systems affect individuals and the social worldIason is a philosopher and Senior Staff Research Scientist at Google DeepMi…
2024 in AI, with Nathan Benaich [not-audio_url] [/not-audio_url]

Duration: 1:48:43
Episode 142Happy holidays! This is one of my favorite episodes of the year — for the third time, Nathan Benaich and I did our yearly roundup of all the AI news and advancements you need to know. This includes selections…
Philip Goff: Panpsychism as a Theory of Consciousness [not-audio_url] [/not-audio_url]

Duration: 1:00:04
Episode 141I spoke with Professor Philip Goff about:* What a “post-Galilean” science of consciousness looks like* How panpsychism helps explain consciousness and the hybrid cosmopsychist viewEnjoy!Philip Goff is a Britis…
Some Changes at The Gradient [not-audio_url] [/not-audio_url]

Duration: 34:25
Hi everyone!If you’re a new subscriber or listener, welcome. If you’re not new, you’ve probably noticed that things have slowed down from us a bit recently. Hugh Zhang, Andrey Kurenkov and I sat down to recap some of The…
Jacob Andreas: Language, Grounding, and World Models [not-audio_url] [/not-audio_url]

Duration: 1:52:43
Episode 140I spoke with Professor Jacob Andreas about:* Language and the world* World models* How he’s developed as a scientistEnjoy!Jacob is an associate professor at MIT in the Department of Electrical Engineering and…
Evan Ratliff: Our Future with Voice Agents [not-audio_url] [/not-audio_url]

Duration: 1:19:59
Episode 139I spoke with Evan Ratliff about:* Shell Game, Evan’s new podcast, where he creates an AI voice clone of himself and sets it loose. * The end of the Longform Podcast and his thoughts on the state of journalism.…
Meredith Ringel Morris: Generative AI's HCI Moment [not-audio_url] [/not-audio_url]

Duration: 1:37:45
Episode 138I spoke with Meredith Morris about:* The intersection of AI and HCI and why we need more cross-pollination between AI and adjacent fields* Disability studies and AI* Generative ghosts and technological determi…
Davidad Dalrymple: Towards Provably Safe AI [not-audio_url] [/not-audio_url]

Duration: 1:20:50
Episode 137I spoke with Davidad Dalrymple about:* His perspectives on AI risk* ARIA (the UK’s Advanced Research and Invention Agency) and its Safeguarded AI ProgrammeEnjoy—and let me know what you think!Davidad is a Prog…
Clive Thompson: Tales of Technology [not-audio_url] [/not-audio_url]

Duration: 2:27:35
Episode 136I spoke with Clive Thompson about:* How he writes* Writing about the climate and biking across the US* Technology culture and persistent debates in AI* PoetryEnjoy—and let me know what you think!Clive is a jou…