Jason Lee
where else to find them
no profile links on file
what they talk about
all 4 in the filters
where they keep going back
no series more than once — all 6 in the filters
on a stage with
never shared a talk
who talks about the same things
nobody close enough yet
| title | conference | date |
|---|---|---|
| Transformers can learn compositional function | The Future of Language Models and Transformers 2025 | 2025-03-31 |
Two concrete results: a shallow single-index teacher-plus-student model shows sharp d^{k-1} plateau emergence times (Hermite coefficients control tiny initial… | ||
| How Transformers Learn Causal Structure with Gradient Descent | Domain Adaptation and Related Areas 2024 | 2024-11-12 |
Gradient descent on a two-layer disentangled transformer with one-hot tokens and softmax attention provably encodes a latent DAG: the first attention layer… | ||
| JOOQ Data Abstractions Without Distraction | Devnexus 2022 | 2022-04-12 |
jOOQ provides a Java DSL that compiles SQL into type-safe queries, with code generation from live DB that creates table/record/DAO classes via a Maven plugin… | ||
| Offline Reinforcement Learning With Only Realizability | Adversarial Approaches in Machine Learning 2022 | 2022-02-17 |
Proposes solving offline RL by solving a regularized primal-dual LP (density-regularized DICE-style objective) to recover an importance weight w and extract a… | ||
| On the Foundations of Deep Learning: SGD, Overparametrization, and Generalization | Frontiers of Deep Learning 2019 | 2019-07-15 |
Experiment with a 50-neuron teacher and 100-neuron learner shows SGD fails on the well-specified 50-neuron model but hits zero test loss every run after… | ||
| Recent Developments in Over-parametrized Neural Networks, Part I | Deep Learning Boot Camp 2019 | 2019-05-28 |
Shows that widening a network often fixes optimization: a realizable two‑layer teacher with 50 neurons resists SGD when fitted with 50 units but reaches… | ||
| Recent Developments in Over-parametrized Neural Networks, Part II | Deep Learning Boot Camp 2019 | 2019-05-28 |
Survey of recent theory connecting wide nets to neural tangent kernel and random features frameworks, with concrete claims: kernel limit arises as width → ∞… | ||
| nothing matches | ||