English
Related papers

Related papers: Does This Gradient Spark Joy?

200 papers

We develop a method for policy architecture search and adaptation via gradient-free optimization which can learn to perform autonomous driving tasks. By learning from both demonstration and environmental reward we develop a model that can…

Machine Learning · Computer Science 2017-10-18 Sayna Ebrahimi , Anna Rohrbach , Trevor Darrell

Recent advancements in Large Reasoning Models (LRMs), exemplified by DeepSeek-R1, have underscored the potential of scaling inference-time compute through Group Relative Policy Optimization (GRPO). However, GRPO frequently suffers from…

Artificial Intelligence · Computer Science 2026-02-09 Yu Zhao , Fan Jiang , Tianle Liu , Bo Zeng , Yu Liu , Longyue Wang , Weihua Luo

Policy gradient methods are very attractive in reinforcement learning due to their model-free nature and convergence guarantees. These methods, however, suffer from high variance in gradient estimation, resulting in poor sample efficiency.…

Machine Learning · Computer Science 2018-11-16 Sergey Pankov

We present a design called Proof of Gradient Optimization (PoGO) for blockchain consensus, where miners produce verifiable evidence of training large-scale machine-learning models. Building on previous work, we incorporate quantized…

Machine Learning · Computer Science 2025-04-24 José I. Orlicki

We study post-training linear autoregressive models with outcome and process rewards. Given a context $\boldsymbol{x}$, the model must predict the response $\boldsymbol{y} \in Y^N$, a sequence of length $N$ that satisfies a $\gamma$ margin…

Machine Learning · Statistics 2026-03-10 Alireza Mousavi-Hosseini , Murat A. Erdogdu

GRPO is a standard approach to endowing pretrained LLMs with reasoning capabilities. It estimates the advantage of an outcome from a group of $K$ outcomes, and promotes those with positive advantages inside a trust region. Since GRPO…

Machine Learning · Computer Science 2026-02-02 Wenzheng Zhang , Karl Stratos

We study the relative power of learning with gradient descent on differentiable models, such as neural networks, versus using the corresponding tangent kernels. We show that under certain conditions, gradient descent achieves small error…

Machine Learning · Computer Science 2021-03-02 Eran Malach , Pritish Kamath , Emmanuel Abbe , Nathan Srebro

Pass@k is a widely used performance metric for verifiable large language model tasks, including mathematical reasoning, code generation, and short-answer reasoning. It defines success if any of $k$ independently sampled solutions passes a…

Machine Learning · Computer Science 2026-02-27 Anas Barakat , Souradip Chakraborty , Khushbu Pahwa , Amrit Singh Bedi

We study two procedures (reverse-mode and forward-mode) for computing the gradient of the validation error with respect to the hyperparameters of any iterative learning algorithm such as stochastic gradient descent. These procedures mirror…

Machine Learning · Statistics 2017-12-13 Luca Franceschi , Michele Donini , Paolo Frasconi , Massimiliano Pontil

Group Relative Policy Optimization (GRPO) is a standard algorithm for reinforcement learning from verifiable rewards, but its group-mean-centered advantage can fail under binary rewards. The failure mode is gradient starvation: when every…

Machine Learning · Computer Science 2026-05-11 Wenhua Nie , Jianan Wu , Junlin Liu , Ziwei Li , Zheng Lin , Zhang Zijian , Yilong Fan , Haoran Zheng , Jyh-Shing Roger Jang

Model-free and model-based reinforcement learning are two ends of a spectrum. Learning a good policy without a dynamic model can be prohibitively expensive. Learning the dynamic model of a system can reduce the cost of learning the policy,…

Robotics · Computer Science 2022-01-19 Arash Mehrjou , Ashkan Soleymani , Stefan Bauer , Bernhard Schölkopf

We consider the well-studied dueling bandit problem, where a learner aims to identify near-optimal actions using pairwise comparisons, under the constraint of differential privacy. We consider a general class of utility-based preference…

Machine Learning · Computer Science 2024-03-25 Aadirupa Saha , Hilal Asi

With the advancement of data-driven techniques, addressing continuous con-trol challenges has become more efficient. However, the reliance of these methods on historical data introduces the potential for unexpected decisions in novel…

Robotics · Computer Science 2023-10-23 Xi Xiong , Lu Liu

In this paper, we implement three state-of-art continuous reinforcement learning algorithms, Deep Deterministic Policy Gradient (DDPG), Proximal Policy Optimization (PPO) and Policy Gradient (PG)in portfolio management. All of them are…

Portfolio Management · Quantitative Finance 2018-11-20 Zhipeng Liang , Hao Chen , Junhao Zhu , Kangkang Jiang , Yanran Li

We derive the optimal differential privacy (DP) parameters of a mechanism that satisfies a given level of R\'enyi differential privacy (RDP). Our result is based on the joint range of two $f$-divergences that underlie the approximate and…

Information Theory · Computer Science 2020-01-17 Shahab Asoodeh , Jiachun Liao , Flavio P. Calmon , Oliver Kosut , Lalitha Sankar

Deep learning models have been extensively adopted in various regions due to their ability to represent hierarchical features, which highly rely on the training set and procedures. Thus, protecting the training process and deep learning…

Cryptography and Security · Computer Science 2025-03-12 Haodi Wang , Tangyu Jiang , Yu Guo , Chengjun Cai , Cong Wang , Xiaohua Jia

In this paper, we consider an infinite horizon average reward Markov Decision Process (MDP). Distinguishing itself from existing works within this context, our approach harnesses the power of the general policy gradient-based algorithm,…

Machine Learning · Computer Science 2024-02-06 Qinbo Bai , Washim Uddin Mondal , Vaneet Aggarwal

Modern machine learning algorithms aim to extract fine-grained information from data to provide accurate predictions, which often conflicts with the goal of privacy protection. This paper addresses the practical and theoretical importance…

Machine Learning · Statistics 2023-07-17 Puyu Wang , Yunwen Lei , Yiming Ying , Ding-Xuan Zhou

Backdoor attacks on deep learning represent a recent threat that has gained significant attention in the research community. Backdoor defenses are mainly based on backdoor inversion, which has been shown to be generic, model-agnostic, and…

Machine Learning · Computer Science 2024-11-11 Xiaoyun Xu , Zhuoran Liu , Stefanos Koffas , Shujian Yu , Stjepan Picek

Policy gradient methods are appealing in deep reinforcement learning but suffer from high variance of gradient estimate. To reduce the variance, the state value function is applied commonly. However, the effect of the state value function…

Machine Learning · Computer Science 2021-08-06 Jiaming Guo , Rui Zhang , Xishan Zhang , Shaohui Peng , Qi Yi , Zidong Du , Xing Hu , Qi Guo , Yunji Chen