Search Results for author:
Found 34 papers, 6 papers with code
Phi-4 Technical Report
We present phi-4, a 14-billion parameter language model developed with a training recipe that is centrally focused on data quality.
Self-Improvement in Language Models: The Sharpening Mechanism
Recent work in language modeling has raised the possibility of self-improvement, where a language models evaluates and refines its own generations to achieve higher performance without external feedback.
ASL STEM Wiki: Dataset and Benchmark for Interpreting STEM Articles
Deaf and hard-of-hearing (DHH) students face significant barriers in accessing science, technology, engineering, and mathematics (STEM) education, notably due to the scarcity of STEM resources in signed languages.
Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
We introduce phi-3-mini, a 3. 8 billion parameter language model trained on 3. 3 trillion tokens, whose overall performance, as measured by both academic benchmarks and internal testing, rivals that of models such as Mixtral 8x7B and GPT-3. 5 (e. g., phi-3-mini achieves 69% on MMLU and 8. 38 on MT-bench), despite being small enough to be deployed on a phone.
Ranked #5 on
MMR total
on MRR-Benchmark
(using extra training data)
Can large language models explore in-context?
We investigate the extent to which contemporary Large Language Models (LLMs) can engage in exploration, a core capability in reinforcement learning and decision making.
Butterfly Effects of SGD Noise: Error Amplification in Behavior Cloning and Autoregression
This work studies training instabilities of behavior cloning with deep neural networks.
Pareto Frontiers in Neural Feature Learning: Data, Compute, Width, and Luck
Finally, we show that the synthetic sparse parity task can be useful as a proxy for real problems requiring axis-aligned feature learning.
Learning Hidden Markov Models Using Conditional Samples
In this paper, we depart from this setup and consider an interactive access model, in which the algorithm can query for samples from the conditional distributions of the HMMs.
Neural Active Learning on Heteroskedastic Distributions
To this end, we demonstrate the catastrophic failure of these active learning algorithms on heteroskedastic distributions and propose a fine-tuning-based approach to mitigate these failures.
Transformers Learn Shortcuts to Automata
Algorithmic reasoning requires capabilities which are most naturally understood through recurrent models of computation, like the Turing machine.
Recurrent Convolutional Neural Networks Learn Succinct Learning Algorithms
For example, on parity problems, the NN learns as well as Gaussian elimination, an efficient algorithm that can be succinctly described.
Hidden Progress in Deep Learning: SGD Learns Parities Near the Computational Limit
There is mounting evidence of emergent phenomena in the capabilities of deep learning methods as we scale up datasets, model sizes, and training times.
Understanding Contrastive Learning Requires Incorporating Inductive Biases
Contrastive learning is a popular form of self-supervised learning that encourages augmentations (views) of the same input to have more similar representations compared to augmentations of different inputs.
Machine Learning for Mechanical Ventilation Control (Extended Abstract)
Mechanical ventilation is one of the most widely used therapies in the ICU.
Anti-Concentrated Confidence Bonuses for Scalable Exploration
Intrinsic rewards play a central role in handling the exploration-exploitation trade-off when designing sequential decision-making algorithms, in both foundational theory and state-of-the-art deep reinforcement learning.
Inductive Biases and Variable Creation in Self-Attention Mechanisms
Self-attention, an architectural motif designed to model long-range interactions in sequential data, has driven numerous recent breakthroughs in natural language processing and beyond.
Sparsity in Partially Controllable Linear Systems
However, in practice, we often encounter systems in which a large set of state variables evolve exogenously and independently of the control inputs; such systems are only partially controllable.
Learning Rate Grafting: Transferability of Optimizer Tuning
In the empirical science of training large neural networks, the learning rate schedule is a notoriously challenging-to-tune hyperparameter, which can depend on all other properties (architecture, optimizer, batch size, dataset, regularization, ...) of the problem.
Acceleration via Fractal Learning Rate Schedules
In practical applications of iterative first-order optimization, the learning rate schedule remains notoriously difficult to understand and expensive to tune.
Deluca -- A Differentiable Control Library: Environments, Methods, and Benchmarking
We present an open-source library of natively differentiable physics and robotics environments, accompanied by gradient-based control methods and a benchmark-ing suite.
Machine Learning for Mechanical Ventilation Control
We consider the problem of controlling an invasive mechanical ventilator for pressure-controlled ventilation: a controller must let air in and out of a sedated patient's lungs according to a trajectory of airway pressures specified by a clinician.
Stochastic Optimization with Laggard Data Pipelines
State-of-the-art optimization is steadily shifting towards massively parallel pipelines with extremely large batch sizes.
Disentangling Adaptive Gradient Methods from Learning Rates
We investigate several confounding factors in the evaluation of optimization algorithms for deep learning.
No-Regret Prediction in Marginally Stable Systems
This requires a refined regret analysis, including a structural lemma showing the current state of the system to be a small linear combination of past states, even if the state grows polynomially.
Revisiting the Generalization of Adaptive Gradient Methods
A commonplace belief in the machine learning community is that using adaptive gradient methods hurts generalization.
Calibration, Entropy Rates, and Memory in Language Models
Building accurate language models that capture meaningful long-term dependencies is a core challenge in natural language processing.
Robust guarantees for learning an autoregressive filter
The optimal predictor for a linear dynamical system (with hidden state and Gaussian noise) takes the form of an autoregressive linear filter, namely the Kalman filter.
Extreme Tensoring for Low-Memory Preconditioning
State-of-the-art models are now trained with billions of parameters, reaching hardware limits in terms of memory consumption.
Efficient Full-Matrix Adaptive Regularization
Due to the large number of parameters of machine learning problems, full-matrix preconditioning methods are prohibitively expensive.
Spectral Filtering for General Linear Dynamical Systems
We give a polynomial-time algorithm for learning latent-state linear dynamical systems without system identification, and without assumptions on the spectral radius of the system's transition matrix.
Towards Provable Control for Unknown Linear Dynamical Systems
We study the control of symmetric linear dynamical systems with unknown dynamics and a hidden state.
Learning Linear Dynamical Systems via Spectral Filtering
We present an efficient and practical algorithm for the online prediction of discrete-time linear dynamical systems with a symmetric transition matrix.
Not-So-Random Features
We propose a principled method for kernel learning, which relies on a Fourier-analytic characterization of translation-invariant or rotation-invariant kernels.
Efficient Regret Minimization in Non-Convex Games
We consider regret minimization in repeated games with non-convex loss functions.
