Seven research papers with AITRICS researchers have been accepted to ICML 2019, thirty-sixth international conference on Machine Learning.

The International Conference on Machine Learning, ICML, is an international academic conference on machine learning. Along with the Conference on Neural Information Processing Systems, it is one of the two primary conferences of high impact in machine learning and artificial intelligence research.

This year, 774 papers were accepted out of 3,424 submissions. AITRICS is among the top corporate research institutions with the most accepted papers, along with IBM, Open AI and Amazon.

Listed below is the overview of AITRICS papers that ICML has accepted.

1. Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks

2. Beyond the Chinese Restaurant and Pitman-Yor processes: Statistical Models with double power-law behavior

3. Trimming the ℓ 1 Regularizer: Statistical Analysis, Optimization, and Applications to Deep Learningr

4. Learning What and Where to Transfer

5. Spectral Approximate Inference

6. Training CNNs with Selective Allocation of Channels

7. Robust Inference via Generative Classifiers for Handling Noisy Labels

Many machine learning tasks such as multiple instance learning, 3D shape recognition, and few-shot image classification are defined on sets of instances. Since solutions to such problems do not depend on the order of elements of the set, models used to address them should be permutation invariant. In our paper, we present an attention-based neural network module, the Set Transformer, specifically designed to model interactions among elements in the input set. The model consists of an encoder and a decoder, both of which rely on attention mechanisms.

In our paper, we present an attention-based neural network module, the Set Transformer, specifically designed to model interactions among elements in the input set. The model consists of an encoder and a decoder, both of which rely on attention mechanisms.

The model consists of an encoder and a decoder, both of which rely on attention mechanisms. In an effort to reduce computational complexity, we introduce an attention scheme inspired by inducing point methods from sparse Gaussian process literature. It reduces the computation time of self-attention from quadratic to linear in the number of elements in the set. We show that our model is theoretically attractive and we evaluate it on a range of tasks, demonstrating the state-of-the-art performance compared to recent methods for set-structured data.