[agents] The AAAI-20 Workshop on Intelligent Process Automation - Call for Papers

[Dell Zhang]

The AAAI-20 Workshop on Intelligent Process Automation (IPA-20)

https://ipa-20.blueprismlabs.com/

TL;DR: A one-day AAAI-20 workshop focusing on learning structured &
executable processes (programs) from human demonstrations, natural
language specifications, or interactions with an environment.
Submissions of previously unpublished or recently published papers
welcome. $1000 Best Paper Award. Free lunch. Interdisciplinary
discussion. Non-archival proceedings.


How to free people from the mundane and repetitive parts of their
daily workload? Robotic Process Automation (RPA) addresses this
problem by developing software agents (robots) that can mimic human
users to perform a variety of business tasks on their computers (e.g.,
processing the invoices received from emails). Since the term RPA was
coined by Blue Prism in 2012, it has spread through enterprises in
many different industrial sections (finance, health,
telecommunication, manufacturing, and so on). It is estimated that the
RPA software market will reach $3.97 billion by 2025.

Current RPA systems are mostly rule-based. Artificial Intelligence
(AI) promises to take RPA to new heights, but so far the AI research
efforts related to the different aspects of RPA have been largely
isolated. This AAAI-20 workshop aims to bridge the gap between the
rapidly growing RPA software industry and the AI research community.
It will bring together researchers from different fields to exchange
ideas and foster discussions on this important, novel application
domain for AI.

# Workshop Topics

Technical topics include, but are not limited to:
- **demo2process** (learning a task-completion software agent from
human demonstrations or behavior logs): *interactive task learning*,
*imitation learning*, *program induction*, *programming by example*,
*process mining*, ...
- **text2process** (learning a task-completion software agent from
step-by-step natural language text descriptions of the process):
*learning by instruction*, *conversational machine learning*, *natural
language programming*, *natural language grounding*, ...
- **task2process** (learning a task-completion software agent directly
from the task as defined by an environment with its reward function or
some input/output examples): *reinforcement learning*, *neural program
synthesis*, *Bayesian program learning*, ...
The common theme is that the learning system's output would not be
simply class labels or numerical predictions, but **structured &
executable processes** (in the form of state-action policies,
if-this-then-that rules, finite-state automatons, or programs in
domain-specific languages), which makes it more challenging than most
of today's machine learning research problems. In addition, such
automated processes must be **safe, robust and explainable** to be
ready for enterprise-level applications, which also poses new research
challenges.

Furthermore, this workshop strongly encourages **interdisciplinary**
submissions from psychologists, economists, or social scientists on
the following topics.
- **human-in-the-loop**: the interaction between human users and
software robots in attended automation.
- **human-outside-the-loop**: the social and organizational impacts of
software robots taking over some workloads or responsibilities from
human users.

# Paper Submission

We welcome contributions of two types.
- New, *previously unpublished* papers which could be either long (6-8
pages) or short (2-4 pages). Such submissions will be given priority.
- Preprints (the accepted version, not the final published version) of
*recently published or in-press* papers (no page limit). Such
submissions must indicate clearly when and where the corresponding
papers have been or will be published.

All submissions should be formatted in the AAAI-20 style and will be
peer-reviewed by multiple reviewers. The authors' identities must be
concealed to enable double-blind peer-review, and all the conflicts of
interests must be declared in advance. The accepted papers will be
published on arXiv.org and they will be included in a **non-archival
workshop proceedings** (which does not prevent the authors from
submitting them to other venues in the future). Blue Prism will
sponsor a **Best Paper Award of $1000**. It will be selected from the
accepted papers by a committee consisting of both academic and
industrial experts.

https://easychair.org/conferences/?conf=ipa20

# Important Dates

- November 8th 2019: Abstract submission due
- November 15th 2019: Full paper submission due
- December 4th 2019: Notification of acceptance/rejection
- December 13th 2019: AAAI-20 early registration deadline
- January 7th 2020: Camera-ready version due
- January 10th 2020: AAAI-20 late registration deadline
- February 7th or 8th 2020: Workshop at AAAI-20

# Workshop Format (Provisional)

One-Day (February 7th or 8th 2020. 9am - 5pm)
- Invited talks
- Oral presentations of contributed papers
- **Free lunch** at Blue Prism's New York office (6 minutes walk from
the conference venue, Hilton New York Midtown)
- Poster presentations of contributed papers
- Panel discussion
- Best paper presentation and ceremony

# Organization Committee

- Dell Zhang (Birkbeck, University of London)
- Andre Freitas (University of Manchester)
- Dacheng Tao (University of Sydney)
- Dawn Song (UC Berkeley)

# Contact

ipa20 at easychair.org