Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userActionable Attribution Maps for Scientific Machine Learning
158
0
0.0
(
0
)
Added by
Shusen Liu
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
The scientific community has been increasingly interested in harnessing the power of deep learning to solve various domain challenges. However, despite the effectiveness in building predictive models, fundamental challenges exist in extracting actionable knowledge from the deep neural network due to their opaque nature. In this work, we propose techniques for exploring the behavior of deep learning models by injecting domain-specific actionable concepts as tunable ``knobs in the analysis pipeline. By incorporating the domain knowledge with generative modeling, we are not only able to better understand the behavior of these black-box models, but also provide scientists with actionable insights that can potentially lead to fundamental discoveries.
rate research
Read More
In this work we explore a new approach for robots to teach themselves about the world simply by observing it. In particular we investigate the effectiveness of learning task-agnostic representations for continuous control tasks. We extend Time-Contrastive Networks (TCN) that learn from visual observations by embedding multiple frames jointly in the embedding space as opposed to a single frame. We show that by doing so, we are now able to encode both position and velocity attributes significantly more accurately. We test the usefulness of this self-supervised approach in a reinforcement learning setting. We show that the representations learned by agents observing themselves take random actions, or other agents perform tasks successfully, can enable the learning of continuous control policies using algorithms like Proximal Policy Optimization (PPO) using only the learned embeddings as input. We also demonstrate significant improvements on the real-world Pouring dataset with a relative error reduction of 39.4% for motion attributes and 11.1% for static attributes compared to the single-frame baseline. Video results are available at https://sites.google.com/view/actionablerepresentations .
Prior gradient-based attribution-map methods rely on handcrafted propagation rules for the non-linear/activation layers during the backward pass, so as to produce gradients of the input and then the attribution map. Despite the promising results achieved, such methods are sensitive to the non-informative high-frequency components and lack adaptability for various models and samples. In this paper, we propose a dedicated method to generate attribution maps that allow us to learn the propagation rules automatically, overcoming the flaws of the handcrafted ones. Specifically, we introduce a learnable plugin module, which enables adaptive propagation rules for each pixel, to the non-linear layers during the backward pass for mask generating. The masked input image is then fed into the model again to obtain new output that can be used as a guidance when combined with the original one. The introduced learnable module can be trained under any auto-grad framework with higher-order differential support. As demonstrated on five datasets and six network architectures, the proposed method yields state-of-the-art results and gives cleaner and more visually plausible attribution maps.
This paper reviews some of the challenges posed by the huge growth of experimental data generated by the new generation of large-scale experiments at UK national facilities at the Rutherford Appleton Laboratory site at Harwell near Oxford. Such Big Scientific Data comes from the Diamond Light Source and Electron Microscopy Facilities, the ISIS Neutron and Muon Facility, and the UKs Central Laser Facility. Increasingly, scientists are now needing to use advanced machine learning and other AI technologies both to automate parts of the data pipeline and also to help find new scientific discoveries in the analysis of their data. For commercially important applications, such as object recognition, natural language processing and automatic translation, deep learning has made dramatic breakthroughs. Googles DeepMind has now also used deep learning technology to develop their AlphaFold tool to make predictions for protein folding. Remarkably, they have been able to achieve some spectacular results for this specific scientific problem. Can deep learning be similarly transformative for other scientific problems? After a brief review of some initial applications of machine learning at the Rutherford Appleton Laboratory, we focus on challenges and opportunities for AI in advancing materials science. Finally, we discuss the importance of developing some realistic machine learning benchmarks using Big Scientific Data coming from a number of different scientific domains. We conclude with some initial examples of our SciML benchmark suite and of the research challenges these benchmarks will enable.
Machine Learning algorithms are good tools for both classification and prediction purposes. These algorithms can further be used for scientific discoveries from the enormous data being collected in our era. We present ways of discovering and understanding astronomical phenomena by applying machine learning algorithms to data collected with radio telescopes. We discuss the use of supervised machine learning algorithms to predict the free parameters of star formation histories and also better understand the relations between the different input and output parameters. We made use of Deep Learning to capture the non-linearity in the parameters. Our models are able to predict with low error rates and give the advantage of predicting in real time once the model has been trained. The other class of machine learning algorithms viz. unsupervised learning can prove to be very useful in finding patterns in the data. We explore how we use such unsupervised techniques on solar radio data to identify patterns and variations, and also link such findings to theories, which help to better understand the nature of the system being studied. We highlight the challenges faced in terms of data size, availability, features, processing ability and importantly, the interpretability of results. As our ability to capture and store data increases, increased use of machine learning to understand the underlying physics in the information captured seems inevitable.
The scientific community has been increasingly interested in harnessing the power of deep learning to solve various domain challenges. However, despite the effectiveness in building predictive models, fundamental challenges exist in extracting actionable knowledge from deep neural networks due to their opaque nature. In this work, we propose techniques for exploring the behavior of deep learning models by injecting domain-specific actionable attributes as tunable knobs in the analysis pipeline. By incorporating the domain knowledge in a generative modeling framework, we are not only able to better understand the behavior of these black-box models, but also provide scientists with actionable insights that can potentially lead to fundamental discoveries.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
