Making Machines Learn from Language Explanations

But humans don’t learn like this. We have a very powerful and intuitive mechanism for communicating information about the world – language!

With just the phrase at least 2 red squares, we’ve summarized the entire dataset presented above in a much more efficient manner.

Language is a crucial medium for human learning: we use it to convey beliefs about the world, teach others, and describe things that are hard to experience directly. Thus, language ought to be a simple and effective way to supervise machine learning models. Yet past approaches to learning from language have struggled to scale up to the general tasks targeted by modern deep learning systems and the freeform language explanations used in these domains. In two short papers presented at ACL 2020 this year, we use deep neural models to learn from language explanations to help tackle a variety of challenging tasks in natural language processing (NLP) and computer vision.

• ExpBERT: Representation Engineering with Natural Language Explanations
• Shaping Visual Representations with Language for Few-shot Classification

What’s the challenge?

Given that language is such an intuitive interface for humans to teach others, why is it so hard to use language for machine learning?

The principal challenge is the grounding problem: understanding language explanations in the context of other inputs. Building models that can understand rich and ambiguous language is tricky enough, but building models that can relate language to the surrounding world is even more challenging. For instance, given the explanation at least two red squares, a model must not only understand the terms red and square, but also how they refer to particular parts of (often complex) inputs.

Past work (1, 2, 3) has relied on semantic parsers which convert natural language statements (e.g. at least two red squares) to formal logical representations (e.g. Count(Square AND Red) > 2). If we can easily check whether explanations apply to our inputs by executing these logical formulas, we can use our explanations as features to train our model. However, semantic parsers only work on simple domains where we can hand-engineer a logical grammar of explanations we might expect to see. They struggle to handle richer and vaguer language or scale up to more complex inputs, such as images.

Fortunately, modern deep neural language models such as BERT are beginning to show promise at solving many language understanding tasks. Our papers propose to alleviate the grounding problem by using neural language models that are either trained to ground language explanations in the domain of interest, or come pre-trained with general-purpose “knowledge” that can be used to interpret explanations. We will show that these neural models allow us to learn from richer and more diverse language for more challenging settings.

Representation Engineering with Natural Language Explanations
In our first paper, we examine how to build text classifiers with language explanations. Consider the task of relation extraction, where we are given a short paragraph and must identify whether two people mentioned in the paragraph are married. While state-of-the-art NLP models can likely solve this task from data alone, humans might use language to describe ways to tell whether two people are married—for example, people who go on honeymoons are typically married. Can such language explanations be used to train better classifiers?

In the same way that we might take an input $x$, and extract features (e.g. the presence of certain words) to train a model, we can use explanations to provide additional features. For example, knowing that honeymoons are relevant for this task, if we can create a honeymoon feature that reliably activates whenever the two people in a paragraph are described as going on a honeymoon, this should be useful signal for training a better model.

But creating such features requires some sort of explanation interpretation mechanism that tells us whether an explanation is true for an input. Semantic parsers are one such tool: given $A$ and $B$ went on honeymoon, we could parse this explanation into a logical form which, when run on an input, produces 1 if the word honeymoon appears between $A$ and $B$. But what about a vaguer explanation like $A$ and $B$ are in love? How can we parse this?

While semantic parsing is efficient and accurate in small domains, it can be overly brittle, as it can only interpret explanations which adhere to a fixed set of grammatical rules and functions that we must specify in advance (e.g. contains and extract_text). Instead, we turn to the soft reasoning capabilities of BERT, a neural language model. BERT is particularly effective at the task of textual entailment: determining whether a sentence implies or contradicts another sentence (e.g. does She ate pizza imply that She ate food? Yes!). In our proposed ExpBERT model, we take a BERT model trained for textual entailment, and instead ask it to identify whether a paragraph in our task entails an explanation. The features produced by BERT during this process replace the indicator features produced by the semantic parser above.

Does the soft reasoning power of BERT improve over semantic parsing? On the marriage identification task, we find that ExpBERT leads to substantial improvements over a classifier that is trained on the input features only (No Explanations). Importantly, using a semantic parser to try to parse explanations doesn’t help much, since there are general explanations (in love) that are difficult to convert to logical forms.