Daniel Wurgaft
@danielwurgaft.bsky.social
PhD @Stanford working w @noahdgoodman and research fellow @GoodfireAI
Studying in-context learning and reasoning in humans and machines
Prev. @UofT CS & Psych
Studying in-context learning and reasoning in humans and machines
Prev. @UofT CS & Psych
Reposted by Daniel Wurgaft
A bias for simplicity by itself does not guarantee good generalization (see the No Free Lunch Theorems). So an inductive bias is only good to the extent that it reflects structure in the data. Is the world simple? The success of deep nets (with their intrinsic Occam's razor) would suggest yes(?)
July 8, 2025 at 1:57 PM
A bias for simplicity by itself does not guarantee good generalization (see the No Free Lunch Theorems). So an inductive bias is only good to the extent that it reflects structure in the data. Is the world simple? The success of deep nets (with their intrinsic Occam's razor) would suggest yes(?)
Hi thanks for the comment! I'm not too familiar with the robot-learning literature but would love to learn more about it!
July 1, 2025 at 7:59 PM
Hi thanks for the comment! I'm not too familiar with the robot-learning literature but would love to learn more about it!
Thank you Andrew!! :)
June 28, 2025 at 11:54 AM
Thank you Andrew!! :)
On a personal note, this is my first full-length first-author paper! @ekdeepl.bsky.social and I both worked so hard on this, and I am so excited about our results and the perspective we bring! Follow for more science of deep learning and human learning!
16/16
16/16
June 28, 2025 at 2:35 AM
On a personal note, this is my first full-length first-author paper! @ekdeepl.bsky.social and I both worked so hard on this, and I am so excited about our results and the perspective we bring! Follow for more science of deep learning and human learning!
16/16
16/16
Thank you to amazing collaborators!
@ekdeepl.bsky.social @corefpark.bsky.social @gautamreddy.bsky.social @hidenori8tanaka.bsky.social @noahdgoodman.bsky.social
See the paper for full results and discussion! And watch for updates! We are working on explaining and unifying more ICL phenomena! 15/
@ekdeepl.bsky.social @corefpark.bsky.social @gautamreddy.bsky.social @hidenori8tanaka.bsky.social @noahdgoodman.bsky.social
See the paper for full results and discussion! And watch for updates! We are working on explaining and unifying more ICL phenomena! 15/
In-Context Learning Strategies Emerge Rationally
Recent work analyzing in-context learning (ICL) has identified a broad set of strategies that describe model behavior in different experimental conditions. We aim to unify these findings by asking why...
arxiv.org
June 28, 2025 at 2:35 AM
Thank you to amazing collaborators!
@ekdeepl.bsky.social @corefpark.bsky.social @gautamreddy.bsky.social @hidenori8tanaka.bsky.social @noahdgoodman.bsky.social
See the paper for full results and discussion! And watch for updates! We are working on explaining and unifying more ICL phenomena! 15/
@ekdeepl.bsky.social @corefpark.bsky.social @gautamreddy.bsky.social @hidenori8tanaka.bsky.social @noahdgoodman.bsky.social
See the paper for full results and discussion! And watch for updates! We are working on explaining and unifying more ICL phenomena! 15/
💡Key takeaways:
3) A top-down, normative perspective offers a powerful, predictive approach for understanding neural networks, complementing bottom-up mechanistic work.
14/
3) A top-down, normative perspective offers a powerful, predictive approach for understanding neural networks, complementing bottom-up mechanistic work.
14/
June 28, 2025 at 2:35 AM
💡Key takeaways:
3) A top-down, normative perspective offers a powerful, predictive approach for understanding neural networks, complementing bottom-up mechanistic work.
14/
3) A top-down, normative perspective offers a powerful, predictive approach for understanding neural networks, complementing bottom-up mechanistic work.
14/
💡Key takeaways:
2) A tradeoff between *loss and complexity* is fundamental to understanding model training dynamics, and gives a unifying explanation for ICL phenomena of transient generalization and task-diversity effects!
13/
2) A tradeoff between *loss and complexity* is fundamental to understanding model training dynamics, and gives a unifying explanation for ICL phenomena of transient generalization and task-diversity effects!
13/
June 28, 2025 at 2:35 AM
💡Key takeaways:
2) A tradeoff between *loss and complexity* is fundamental to understanding model training dynamics, and gives a unifying explanation for ICL phenomena of transient generalization and task-diversity effects!
13/
2) A tradeoff between *loss and complexity* is fundamental to understanding model training dynamics, and gives a unifying explanation for ICL phenomena of transient generalization and task-diversity effects!
13/
💡Key takeaways:
1) Is ICL Bayes-optimal? We argue the better question is *under what assumptions*. Cautiously, we conclude that ICL can be seen as approx. Bayesian under a simplicity bias and sublinear sample efficiency (though see our appendix for an interesting deviation!)
12/
1) Is ICL Bayes-optimal? We argue the better question is *under what assumptions*. Cautiously, we conclude that ICL can be seen as approx. Bayesian under a simplicity bias and sublinear sample efficiency (though see our appendix for an interesting deviation!)
12/
June 28, 2025 at 2:35 AM
💡Key takeaways:
1) Is ICL Bayes-optimal? We argue the better question is *under what assumptions*. Cautiously, we conclude that ICL can be seen as approx. Bayesian under a simplicity bias and sublinear sample efficiency (though see our appendix for an interesting deviation!)
12/
1) Is ICL Bayes-optimal? We argue the better question is *under what assumptions*. Cautiously, we conclude that ICL can be seen as approx. Bayesian under a simplicity bias and sublinear sample efficiency (though see our appendix for an interesting deviation!)
12/
Ablations of our analytical expression show the modeled computational constraints, in their assumed functional forms, are crucial!
11/
11/
June 28, 2025 at 2:35 AM
Ablations of our analytical expression show the modeled computational constraints, in their assumed functional forms, are crucial!
11/
11/
And reveals some interesting findings: MLP width increases memorization, which is captured by our model as a reduced simplicity bias!
10/
10/
June 28, 2025 at 2:35 AM
And reveals some interesting findings: MLP width increases memorization, which is captured by our model as a reduced simplicity bias!
10/
10/
Our framework also makes novel Predictions:
🔹**Sub-linear** sample efficiency → sigmoidal transition from generalization to memorization
🔹**Rapid** behavior change near the M–G crossover boundary
🔹**Superlinear** scaling of time to transience as data diversity increases
9/
🔹**Sub-linear** sample efficiency → sigmoidal transition from generalization to memorization
🔹**Rapid** behavior change near the M–G crossover boundary
🔹**Superlinear** scaling of time to transience as data diversity increases
9/
June 28, 2025 at 2:35 AM
Our framework also makes novel Predictions:
🔹**Sub-linear** sample efficiency → sigmoidal transition from generalization to memorization
🔹**Rapid** behavior change near the M–G crossover boundary
🔹**Superlinear** scaling of time to transience as data diversity increases
9/
🔹**Sub-linear** sample efficiency → sigmoidal transition from generalization to memorization
🔹**Rapid** behavior change near the M–G crossover boundary
🔹**Superlinear** scaling of time to transience as data diversity increases
9/
Intuitively, what does this predictive account imply? A rational tradeoff between a strategy's loss and complexity!
🔵Early: A simplicity bias (prior) favors a less complex strategy (G)
🔴Late: reducing loss (likelihood) favors a better-fitting, but more complex strategy (M)
8/
🔵Early: A simplicity bias (prior) favors a less complex strategy (G)
🔴Late: reducing loss (likelihood) favors a better-fitting, but more complex strategy (M)
8/
June 28, 2025 at 2:35 AM
Intuitively, what does this predictive account imply? A rational tradeoff between a strategy's loss and complexity!
🔵Early: A simplicity bias (prior) favors a less complex strategy (G)
🔴Late: reducing loss (likelihood) favors a better-fitting, but more complex strategy (M)
8/
🔵Early: A simplicity bias (prior) favors a less complex strategy (G)
🔴Late: reducing loss (likelihood) favors a better-fitting, but more complex strategy (M)
8/
Fitting the three free parameters of our expression, we see that across checkpoints from 11 different runs, we almost perfectly predict *next-token predictions* and the relative distance maps!
We now have a predictive model of task diversity effects and transience!
7/
We now have a predictive model of task diversity effects and transience!
7/
June 28, 2025 at 2:35 AM
Fitting the three free parameters of our expression, we see that across checkpoints from 11 different runs, we almost perfectly predict *next-token predictions* and the relative distance maps!
We now have a predictive model of task diversity effects and transience!
7/
We now have a predictive model of task diversity effects and transience!
7/