Machine Learning Blog | ML@CMU | Carnegie Mellon University

machine learning Research

RL for Planning and Planning for RL

by Benjamin Eysenbach / February 13, 2020

The figure above illustrates the method:(a) Goal-conditioned RL often fails to reach distant goals, but can successfully reach the goal if starting nearby (inside the green region). (b) Our goal is to use observations in our replay buffer (yellow squares) as waypoints leading to the goal. (c) We automatically find these waypoints by using the agent’s value function to predict when two states are nearby, and building the corresponding graph. (d) We run graph search to find the sequence of…

14 2507
Read More

machine learning Research

RoboNet: A Dataset for Large-Scale Multi-Robot Learning

November 26, 2019

In the last decade, we’ve seen learning-based systems provide transformative solutions for a wide range of perception and reasoning problems, from recognizing objects in images to recognizing and translating human speech. Recent progress in deep reinforcement learning (i.e. integrating deep…

40 4365

machine learning Research

ARUBA: Learning-to-Learn with Less Regret

November 22, 2019

In the classical machine learning setup, we aim to learn a single model for a single task given many training samples from the same distribution. However, in many practical applications, we are in fact exposed to several distinct yet related…

41 3962

Educational federated learning machine learning

Federated Learning: Challenges, Methods, and Future Directions

November 12, 2019

What is federated learning? How does it differ from traditional large-scale machine learning, distributed optimization, and privacy-preserving data analysis? What do we understand currently about federated learning, and what problems are left to explore? In this post, we briefly answer…

38 9614

machine learning Research

Path Length Bounds for Gradient Descent

October 25, 2019

In today's post, we will discuss an interesting property concerning the trajectory of gradient descent iterates, namely the length of the Gradient Descent curve. Let us assume we want to minimize the function shown in Figure 1 starting from a…

44 4502

deep learning machine learning Research

Ultra-Wide Deep Nets and the Neural Tangent Kernel (NTK)

October 3, 2019

Traditional wisdom in machine learning holds that there is a careful trade-off between training error and generalization gap. There is a "sweet spot" for the model complexity such that the model (i) is big enough to achieve reasonably good training…

66 15452 2

machine learning Research

On Learning Invariant Representations for Domain Adaptation

September 13, 2019

One of the backbone assumptions underpinning the generalization theory of supervised learning algorithms is that the test distribution should be the same as the training distribution. However in many real-world applications it is usually time-consuming or even infeasible to collect…

56 7226

machine learning Research

Carnegie Mellon University, Accepted Papers at NeurIPS 2019

September 6, 2019

We are proud to present the following papers at the 33rd Conference on Neural Information Processing Systems (NeurIPS) in Vancouver, Canada. Check back for an update with poster numbers and links once the camera-ready papers become available. If you are…

66 8198

artificial intelligence FATE learning theory reinforcement learning Research

Policy Certificates and Minimax-Optimal PAC Bounds for Episodic Reinforcement Learning

August 16, 2019

Designing reinforcement learning methods which find a good policy with as few samples as possible is a key goal of both empirical and theoretical research. On the theoretical side there are two main ways, regret- or PAC (probably approximately correct)…

72 2884

machine learning Research

Regret Circuits: Composability of Regret Minimizers

August 2, 2019

Automated decision-making is one of the core objectives of artificial intelligence. Not surprisingly, over the past few years, entire new research fields have emerged to tackle that task. This blog post is concerned with regret minimization, one of the central…

78 12331

Older Posts

Machine Learning Blog | ML@CMU | Carnegie Mellon University

Statistics:

Categories: