Lower Bounds for Passive and Active Learning

نویسندگان

  • Maxim Raginsky
  • Alexander Rakhlin
چکیده

We develop unified information-theoretic machinery for deriving lower bounds for passive and active learning schemes. Our bounds involve the so-called Alexander’s capacity function. The supremum of this function has been recently rediscovered by Hanneke in the context of active learning under the name of “disagreement coefficient.” For passive learning, our lower bounds match the upper bounds of Giné and Koltchinskii up to constants and generalize analogous results of Massart and Nédélec. For active learning, we provide first known lower bounds based on the capacity function rather than the disagreement coefficient.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Comparison of Learning and Memory in Morphine Dependent Rats using Different Behavioral Models

There are several conflicting evidences showing the effect of morphine on learning and memory processes. In the present study the effect of chronic morphine administration on passive avoidance, active avoidance and spatial learning and memory of morphine dependent male rats using Passive Avoidance shuttle box and Morris Water Maze tasks were investigated, respectively. Male rats received morphi...

متن کامل

Comparison of Learning and Memory in Morphine Dependent Rats using Different Behavioral Models

There are several conflicting evidences showing the effect of morphine on learning and memory processes. In the present study the effect of chronic morphine administration on passive avoidance, active avoidance and spatial learning and memory of morphine dependent male rats using Passive Avoidance shuttle box and Morris Water Maze tasks were investigated, respectively. Male rats received morphi...

متن کامل

Lower Bounds on Active Learning for Graphical Model Selection

We consider the problem of estimating the underlying graph associated with a Markov random field, with the added twist that the decoding algorithm can iteratively choose which subsets of nodes to sample based on the previous samples, resulting in an active learning setting. Considering both Ising and Gaussian models, we provide algorithm-independent lower bounds for high-probability recovery wi...

متن کامل

The Effects of Cooperative Language Learning Strategies on Learning Active and Passive Structures among Iranian EFL Learners

This study aims at investigating the effects of cooperative language learning on learning active and passive structures among Iranian EFL students. The participants of the study were 60 high school students that were selected from third grade of Barikbin high school in Qazvin. All of the participants were male. Their level of proficiency was intermediate. Then the participants were divided into...

متن کامل

Upper and Lower Error Bounds for Active Learning

This paper analyzes the potential advantages and theoretical challenges of ”active learning” algorithms. Active learning involves sequential, adaptive sampling procedures that use information gleaned from previous samples in order to focus the sampling and accelerate the learning process relative to “passive learning” algorithms, which are based on non-adaptive (usually random) samples. There a...

متن کامل

ذخیره 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 نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2011