Merge pull request #7 from aidos-lab/es-dev

Adding our toolbox! 🛠️

Author: Pseudomanifold

content/_index.md

@@ -2,7 +2,7 @@
 title: "AIDOS Lab"
 ---
 
-{{< figure src="logo_hexagon.svg" width="128" class="center" alt="AIDOS Lab Logo" >}}
+<!-- {{< figure src="logo_hexagon.svg" width="128" class="center" alt="AIDOS Lab Logo" >}} -->
 
 # Welcome
 
@@ -12,20 +12,27 @@ We are fascinated by
 discovering hidden structures in complex data sets, in particular those
 arising in healthcare applications.
 
+{{< figure src="new_logo_standin.svg" width="600" class="center" alt="New Lab Logo" >}}
+
 Our primary research interests are situated at the intersection of
-**geometry, topology, and machine learning**.
-We want to make use of geometrical and
-topological information to imbue
-neural networks with more information in their respective tasks, leading
-to better and more robust outcomes.
-Along the way, we develop new **manifold learning** techniques, new
+**geometry, topology, and machine learning**. From data-centric evaluation frameworks to novel architecture developments, we make use of geometrical and
+topological information to imbue our methods with foundations that foster interpretability and robustness. Along the way, we develop new **manifold learning** techniques, new
 **representation learning** algorithms, and much more.
 
 Following the dictum 'theory without practice is empty,' we
 address challenges in biomedicine and healthcare
-applications.
+applications. Check out our [research page]({{< relref "/research" >}}) to learn more about our current projects.
+
+## Mission Statement
 
-Check out [our research to learn more]({{< relref "/research" >}}).
+'AIDOS' has two meanings that complement each other well. The first
+meaning refers to our mission statement, viz. to develop **A**rtificial
+**I**ntelligence for **D**iscovering **O**bscured **S**hapes. The
+second meaning originates from the Greek word 'αἰδώς,' which means
+'awe,' 'reverence,' or 'humility.' This awe or humility should serve as
+one of our guiding principles when we work on challenging problems in
+healthcare research, aiming to improve our world using machine
+learning.
 
 ## Mission Statement
 

content/research.md

@@ -4,6 +4,19 @@ title: "Research"
 
 # Research
 
+The AIDOS lab is dedicated to establishing foundational principles in machine learning. Leveraging our experience in computational geometry and topology, we focus on shaping well-principled methods to address holes in rapidly evolving AI landscape. We see ourselves as toolsmiths, crafting both observational and interventional frameworks using concepts such as the Euler characteristic, metric space magnitude, curvature, persistent homology, etc. Whether working with graphs, images, or natural language, our goal is to build tools that tackle the most difficult questions, prioritizing simplicity, elegance, and interpretability over mere performance. We hope our work can give back to the community, empowering new research directions and application developments grounded in principled methods.
+
+# Toolbox
+Here is a collection of tools that have been developed by the AIDOS Lab, in order from most to least recent.
+{{< tool "scott" >}}
+{{< tool "magnipy" >}}
+{{< tool "presto" >}}
+{{< tool "dect" >}}
+{{< tool "tardis" >}}
+{{< tool "orchid" >}}
+
+# Publications
+
 Here are all publications of lab members, sorted by year. Publications
 appear in the order in which they are accepted.
 

content/tools.md

@@ -0,0 +1,15 @@
+# Toolbox
+
+Here is a collection of tools that have been developed by the AIDOS Lab, in order from most to least recent.
+
+{{< tool "scott" >}}
+
+{{< tool "magnipy" >}}
+
+{{< tool "presto" >}}
+
+{{< tool "dect" >}}
+
+{{< tool "orchid" >}}
+
+{{< tool "tardis" >}}

data/tools/TARDIS.toml

@@ -0,0 +1,15 @@
+name = "TARDIS"
+paper   = "Topological Algorithms for Robust DIscovery of Singularities"
+image   = "tardis.svg"
+venue   = "ICML 2023"
+description = """\
+The manifold hypothesis, which assumes that data lies on or close to an unknown manifold of 
+low intrinsic dimension, is a staple of modern machine learning research. However, recent 
+work has shown that real-world data exhibits distinct non-manifold structures, i.e. singularities, 
+that can lead to erroneous findings. We address this issue by developing a topological framework that 
+(i) quantifies the local intrinsic dimension, and (ii) yields a Euclidicity score for assessing the 
+'manifoldness' of a point along multiple scales.
+"""
+arXiv   = "2210.00069"
+pub     = "https://arxiv.org/abs/2210.00069"
+repo    = "https://github.com/aidos-lab/TARDIS"

data/tools/dect.toml

@@ -0,0 +1,15 @@
+name = "DECT"
+paper   = "Differentiable Euler Characteristic Transform"
+image   = "dect.png"
+venue   = "ICLR 2024"
+description = """\
+The Euler Characteristic Transform (ECT) has proven to be a powerful representation, 
+combining geometrical and topological characteristics of shapes and graphs.
+With the Differentiable Euler Characteristic Transform (DECT), we provide a fast 
+and computationally efficient implementation of a differentiable, end-to-end-trainable ECT, 
+which can be integrated into deep neural networks.
+"""
+arXiv   = "2310.07630"
+pub     = "https://arxiv.org/abs/2310.07630"
+repo    = "https://github.com/aidos-lab/dect"
+pip     = "pip install git+https://github.com/aidos-lab/dect.git"

data/tools/magnipy.toml

@@ -0,0 +1,13 @@
+name = "magnipy"
+paper   = "Metric Space Magnitude Computations"
+image   = "magnipy.svg"
+venue   = "NeurIPS 2024"
+description = """\
+The magnitude of a metric space is a powerful invariant that provides a measure of the 'effective size' 
+of a space across multiple scales, while also capturing numerous geometrical properties, 
+such as curvature, density, or entropy. We provide a toolbox for computing and comparing
+the magnitude of metric spaces.  
+"""
+arXiv   = "2311.16054"
+pub     = "https://arxiv.org/abs/2311.16054"
+repo    = "https://github.com/aidos-lab/magnipy"

data/tools/orchid.toml

@@ -0,0 +1,16 @@
+name   = "Orchid"
+paper = "Ollivier–Ricci Curvature for Hypergraphs: A Unified Framework"
+image   = "orchid.png"
+venue   = "ICLR 2023"
+description = """\
+Bridging geometry and topology, curvature is a powerful and expressive
+invariant. While the utility of curvature has been theoretically and
+empirically confirmed in the context of manifolds and graphs, its
+generalisation to the emerging domain of hypergraphs has remained
+largely unexplored. Our toolbox aims to fill this gap and presents a new
+framework for hypergraph curvature.
+"""
+arXiv   = "2210.12048"
+pub     = "https://arxiv.org/abs/2210.12048"
+repo    = "https://github.com/aidos-lab/orchid"
+pip     = "julia -e 'using Pkg; Pkg.add(url='https://github.com/aidos-lab/orchid.git')'"

data/tools/presto.toml

@@ -0,0 +1,9 @@
+name    = "PRESTO"
+paper   = "Mapping the Multiverse of Latent Representations"
+image   = "presto.svg"
+description     = """\
+            The world of machine learning research is riddled with small decisions, from data collection, cleaning, into model selection and parameter tuning 🎶. Each combination of data, implementation, and modeling decisions leads to a potential universe where we can analyze and interpret results. Together, these form a multiverse! 🌌 With PRESTO (Projected Embedding Similarity based on Topological Overlays), we offer topological tools to efficiently measure the structural variation between representations that arise from different choices in a machine learning workflow.
+          """
+repo    = "https://github.com/aidos-lab/Presto"
+pub     = "https://doi.org/10.48550/arXiv.2402.01514"
+pip     = "pip install presto-multiverse"

data/tools/scott.toml

@@ -0,0 +1,10 @@
+name    = "SCOTT"
+paper   = "Curvature Filtrations for Graph Generative Model Evaluation"
+image   = "scott.jpg"
+description     = """\
+          SCOTT (Synthesizing Curvature Operations and Topological Tools) is a Python package for computing **curvature filtrations** for graphs and graph distributions. Our method introduces a novel way to compare graph distributions by combining discrete curvature on graphs with persistent homology, providing descriptors of graph sets that are: *robust*, *stable*, *expressive*, and *compatible with statistical testing*. The package is highly adaptable, offering several options for user customization, including different curvature computation methods and diverse metrics for comparing persistent homology outputs.
+          """
+repo    = "https://github.com/aidos-lab/curvature-filtrations"
+pub     = "https://doi.org/10.48550/arXiv.2301.12906"
+pip     = "pip install curvature-filtrations"
+

layouts/partials/codeblock.html

@@ -0,0 +1,6 @@
+<dt>
+    <span class="card-category">{{ .name }}</span>
+</dt>
+<dd>
+    <span style="font-family: monospace;">{{ .text }}</span>
+</dd>