On Learning Latent Models with Multi-Instance Weak Supervision

by   Kaifu Wang, et al.

We consider a weakly supervised learning scenario where the supervision signal is generated by a transition function σ of labels associated with multiple input instances. We formulate this problem as multi-instance Partial Label Learning (multi-instance PLL), which is an extension to the standard PLL problem. Our problem is met in different fields, including latent structural learning and neuro-symbolic integration. Despite the existence of many learning techniques, limited theoretical analysis has been dedicated to this problem. In this paper, we provide the first theoretical study of multi-instance PLL with possibly an unknown transition σ. Our main contributions are as follows. Firstly, we propose a necessary and sufficient condition for the learnability of the problem. This condition non-trivially generalizes and relaxes the existing small ambiguity degree in the PLL literature, since we allow the transition to be deterministic. Secondly, we derive Rademacher-style error bounds based on a top-k surrogate loss that is widely used in the neuro-symbolic literature. Furthermore, we conclude with empirical experiments for learning under unknown transitions. The empirical results align with our theoretical findings; however, they also expose the issue of scalability in the weak supervision literature.


page 1

page 2

page 3

page 4


Learnability with Indirect Supervision Signals

Learning from indirect supervision signals is important in real-world AI...

Leveraged Weighted Loss for Partial Label Learning

As an important branch of weakly supervised learning, partial label lear...

Disambiguation of weak supervision with exponential convergence rates

Machine learning approached through supervised learning requires expensi...

mil-benchmarks: Standardized Evaluation of Deep Multiple-Instance Learning Techniques

Multiple-instance learning is a subset of weakly supervised learning whe...

Lower-bounded proper losses for weakly supervised classification

This paper discusses the problem of weakly supervised learning of classi...

Towards Learning Causal Representations from Multi-Instance Bags

Although humans can easily identify the object of interest from groups o...

Please sign up or login with your details

Forgot password? Click here to reset