The Interplay Between Stability and Regret in Online Learning
نویسندگان
چکیده
This paper considers the stability of online learning algorithms and its implications for learnability (bounded regret). We introduce a novel quantity called forward regret that intuitively measures how good an online learning algorithm is if it is allowed a one-step look-ahead into the future. We show that given stability, bounded forward regret is equivalent to bounded regret. We also show that the existence of an algorithm with bounded regret implies the existence of a stable algorithm with bounded regret and bounded forward regret. The equivalence results apply to general, possibly non-convex problems. To the best of our knowledge, our analysis provides the first general connection between stability and regret in the online setting that is not restricted to a particular class of algorithms. Our stability-regret connection provides a simple recipe for analyzing regret incurred by any online learning algorithm. Using our framework, we analyze several existing online learning algorithms as well as the “approximate” versions of algorithms like RDA that solve an optimization problem at each iteration. Our proofs are simpler than existing analysis for the respective algorithms, show a clear trade-off between stability and forward regret, and provide tighter regret bounds in some cases. Furthermore, using our recipe, we analyze “approximate” versions of several algorithms such as follow-the-regularized-leader (FTRL) that requires solving an optimization problem at each step.
منابع مشابه
Stability Conditions for Online Learnability
Stability is a general notion that quantifies the sensitivity of a learning algorithm’s output to small change in the training dataset (e.g. deletion or replacement of a single training sample). Such conditions have recently been shown to be more powerful to characterize learnability in the general learning setting under i.i.d. samples where uniform convergence is not necessary for learnability...
متن کاملThe Interplay between Young Learners' Sense of Self-Efficacy in Reading Comprehension and English Language Proficiency
This study intended to explore the interplay between young language learners' sense of self-efficacy regarding reading comprehension in their reading test performance associated with learning English among universities. To undertake the study, a purposive sampling method was adopted. A total of 60 freshmen undergraduate learners of English consented to participate in this study. A self-efficac...
متن کاملUnified Algorithms for Online Learning and Competitive Analysis
Online learning and competitive analysis are two widely studied frameworks for online decisionmaking settings. Despite the frequent similarity of the problems they study, there are significant differences in their assumptions, goals and techniques, hindering a unified analysis and richer interplay between the two. In this paper, we provide several contributions in this direction. We provide a s...
متن کاملThe Interplay between Ethnic Identities and Social Attitude toward Foreign Language Learning and Language Proficiency of Young Gilak EFL Learners
As a social-psychological phenomenon, language learning involves several factors. The two significant factors that attracted scholars’ attention recently are ethnicity and social attitude toward L2. Taking in to account this issue, the present study sought to investigate the relationship between Gilak ethnic identity, social attitude toward foreign language, and L2 proficiency...
متن کاملOn the Interplay of Self-Esteem, Proficiency Level, and Language Learning Strategies Among Iranian L2 Learners
It is axiomatic that L2 teaching and learning is a process that requires dynamic involvement of L2 learners in the acquisition of knowledge and skills. L2 learners need to be assisted in setting individual learning goals. They should also be given the exposure to and guidance in effective language learning strategies (LLSs) in order to build a high level of confidence in the learning process. T...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- CoRR
دوره abs/1211.6158 شماره
صفحات -
تاریخ انتشار 2012