Two lines of work are taking the central stage in AI research. On the on...
Disentanglement aims to recover meaningful latent ground-truth factors f...
In-context learning, a capability that enables a model to learn from inp...
Agents that can understand and reason over the dynamics of objects can h...
Foundation models are redefining how AI systems are built. Practitioners...
Modern deep learning systems are fragile and do not generalize well unde...
The theory of representation learning aims to build methods that provabl...
Federated learning aims to train predictive models for data that is
dist...
Humans have a remarkable ability to disentangle complex sensory inputs (...
Machine learning models often fail to generalize well under distribution...
Locally interpretable model agnostic explanations (LIME) method is one o...
A key goal of unsupervised representation learning is "inverting" a data...
Treatment effect estimation from observational data is a fundamental pro...
The invariance principle from causality is at the heart of notable appro...
Can models with particular structure avoid being biased towards spurious...
A major bottleneck in the real-world applications of machine learning mo...
Inferring causal individual treatment effect (ITE) from observational da...
Non-convex optimization problems are challenging to solve; the success a...
Recently, invariant risk minimization (IRM) was proposed as a promising
...
Recently, invariant risk minimization (IRM) (Arjovsky et al.) was propos...
The standard risk minimization paradigm of machine learning is brittle w...
Recently, a method called the Mutual Information Neural Estimator (MINE)...
A clinician desires to use a risk-stratification method that achieves
co...
Existing metrics in competing risks survival analysis such as concordanc...
Machine Learning models have proved extremely successful for a wide vari...