Search Results for author:
Found 12 papers, 0 papers with code
LiMAML: Personalization of Deep Recommender Models via Meta Learning
In the realm of recommender systems, the ubiquitous adoption of deep neural networks has emerged as a dominant paradigm for modeling diverse business objectives.
LiRank: Industrial Large Scale Ranking Models at LinkedIn
We present LiRank, a large-scale ranking framework at LinkedIn that brings to production state-of-the-art modeling architectures and optimization methods.
A Precise Characterization of SGD Stability Using Loss Surface Geometry
In this paper, we delve deeper into the relationship between linear stability and sharpness.
MultiSlot ReRanker: A Generic Model-based Re-Ranking Framework in Recommendation Systems
In this paper, we propose a generic model-based re-ranking framework, MultiSlot ReRanker, which simultaneously optimizes relevance, diversity, and freshness.
FFSplit: Split Feed-Forward Network For Optimizing Accuracy-Efficiency Trade-off in Language Model Inference
This usually results in a trade-off between model accuracy and efficiency.
QuantEase: Optimization-based Quantization for Language Models
Particularly noteworthy is our outlier-aware algorithm's capability to achieve near or sub-3-bit quantization of LLMs with an acceptable drop in accuracy, obviating the need for non-uniform quantization or grouping techniques, improving upon methods such as SpQR by up to two times in terms of perplexity.
mSAM: Micro-Batch-Averaged Sharpness-Aware Minimization
Modern deep learning models are over-parameterized, where different optima can result in widely varying generalization performance.
Improved Deep Neural Network Generalization Using m-Sharpness-Aware Minimization
To that end, this paper presents a thorough empirical evaluation of mSAM on various tasks and datasets.
Heterogeneous Calibration: A post-hoc model-agnostic framework for improved generalization
We introduce the notion of heterogeneous calibration that applies a post-hoc model-agnostic transformation to model outputs for improving AUC performance on binary classification tasks.
Logit Attenuating Weight Normalization
Over-parameterized deep networks trained using gradient-based optimizers are a popular choice for solving classification and ranking problems.
Transitioning from Real to Synthetic data: Quantifying the bias in model
This study aims to establish a trade-off between bias and fairness in the models trained using synthetic data.
Smoothed Gaussian Mixture Models for Video Classification and Recommendation
SGMM represents each video by the parameters of a Gaussian mixture model (GMM) trained for that video.
