• Publications
  • Influence
Scalable Kernel Methods via Doubly Stochastic Gradients
TLDR
An approach that scales up kernel methods using a novel concept called "doubly stochastic functional gradients" based on the fact that many kernel methods can be expressed as convex optimization problems, which can readily scale kernel methods up to the regimes which are dominated by neural nets.
Margin Based Active Learning
TLDR
The effectiveness of the framework for margin based active learning of linear separators both in the realizable case and in a specific noisy setting related to the Tsybakov small noise condition is analyzed.
Approximate clustering without the approximation
TLDR
If any c-approximation to the given clustering objective φ is e-close to the target, then this paper shows that this guarantee can be achieved for any constant c > 1, and for the min-sum objective the authors can do this for any Constant c > 2.
The Power of Localization for Efficiently Learning Linear Separators with Noise
TLDR
This work provides the first polynomial-time active learning algorithm for learning linear separators in the presence of malicious noise or adversarial label noise, and achieves a label complexity whose dependence on the error parameter ϵ is polylogarithmic (and thus exponentially better than that of any passive algorithm).
Improved Guarantees for Learning via Similarity Functions
TLDR
A new notion of a “good similarity function” is provided that builds upon the previous definition of Balcan and Blum (2006) but improves on it in two important ways, and it is proved that for distribution-specific PAC learning, the new notion is strictly more powerful than the traditional notion ofA large-margin kernel.
On a theory of learning with similarity functions
TLDR
This work develops an alternative, more general theory of learning with similarity functions (i.e., sufficient conditions for a similarity function to allow one to learn well) that does not require reference to implicit spaces, anddoes not require the function to be positive semi-definite (or even symmetric).
Co-Training and Expansion: Towards Bridging Theory and Practice
TLDR
A much weaker "expansion" assumption on the underlying data distribution is proposed, that is proved to be sufficient for iterative co-training to succeed given appropriately strong PAC-learning algorithms on each feature set, and that to some extent is necessary as well.
Approximation algorithms and online mechanisms for item pricing
TLDR
The approximation algorithms can be fed into the generic reduction of Balcan et al.
Clustering under Perturbation Resilience
TLDR
This paper presents an algorithm that can optimally cluster instances resilient to $(1 + \sqrt{2})$-factor perturbations, solving an open problem of Awasthi et al.
...
1
2
3
4
5
...